diff options
Diffstat (limited to 'lib/htmlpurifier/library')
350 files changed, 25637 insertions, 0 deletions
diff --git a/lib/htmlpurifier/library/HTMLPurifier.auto.php b/lib/htmlpurifier/library/HTMLPurifier.auto.php new file mode 100644 index 000000000..1960c399f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.auto.php @@ -0,0 +1,11 @@ +<?php + +/** + * This is a stub include that automatically configures the include path. + */ + +set_include_path(dirname(__FILE__) . PATH_SEPARATOR . get_include_path() ); +require_once 'HTMLPurifier/Bootstrap.php'; +require_once 'HTMLPurifier.autoload.php'; + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.autoload.php b/lib/htmlpurifier/library/HTMLPurifier.autoload.php new file mode 100644 index 000000000..62da5b60d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.autoload.php @@ -0,0 +1,26 @@ +<?php + +/** + * @file + * Convenience file that registers autoload handler for HTML Purifier. + * It also does some sanity checks. + */ + +if (function_exists('spl_autoload_register') && function_exists('spl_autoload_unregister')) { + // We need unregister for our pre-registering functionality + HTMLPurifier_Bootstrap::registerAutoload(); + if (function_exists('__autoload')) { + // Be polite and ensure that userland autoload gets retained + spl_autoload_register('__autoload'); + } +} elseif (!function_exists('__autoload')) { + function __autoload($class) { + return HTMLPurifier_Bootstrap::autoload($class); + } +} + +if (ini_get('zend.ze1_compatibility_mode')) { + trigger_error("HTML Purifier is not compatible with zend.ze1_compatibility_mode; please turn it off", E_USER_ERROR); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.func.php b/lib/htmlpurifier/library/HTMLPurifier.func.php new file mode 100644 index 000000000..56a55b2fe --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.func.php @@ -0,0 +1,23 @@ +<?php + +/** + * @file + * Defines a function wrapper for HTML Purifier for quick use. + * @note ''HTMLPurifier()'' is NOT the same as ''new HTMLPurifier()'' + */ + +/** + * Purify HTML. + * @param $html String HTML to purify + * @param $config Configuration to use, can be any value accepted by + * HTMLPurifier_Config::create() + */ +function HTMLPurifier($html, $config = null) { + static $purifier = false; + if (!$purifier) { + $purifier = new HTMLPurifier(); + } + return $purifier->purify($html, $config); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.includes.php b/lib/htmlpurifier/library/HTMLPurifier.includes.php new file mode 100644 index 000000000..0ceff6a92 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.includes.php @@ -0,0 +1,221 @@ +<?php + +/** + * @file + * This file was auto-generated by generate-includes.php and includes all of + * the core files required by HTML Purifier. Use this if performance is a + * 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 4.4.0 + * + * @warning + * You must *not* include any other HTML Purifier files before this file, + * because 'require' not 'require_once' is used. + * + * @warning + * This file requires that the include path contains the HTML Purifier + * library directory; this is not auto-set. + */ + +require 'HTMLPurifier.php'; +require 'HTMLPurifier/AttrCollections.php'; +require 'HTMLPurifier/AttrDef.php'; +require 'HTMLPurifier/AttrTransform.php'; +require 'HTMLPurifier/AttrTypes.php'; +require 'HTMLPurifier/AttrValidator.php'; +require 'HTMLPurifier/Bootstrap.php'; +require 'HTMLPurifier/Definition.php'; +require 'HTMLPurifier/CSSDefinition.php'; +require 'HTMLPurifier/ChildDef.php'; +require 'HTMLPurifier/Config.php'; +require 'HTMLPurifier/ConfigSchema.php'; +require 'HTMLPurifier/ContentSets.php'; +require 'HTMLPurifier/Context.php'; +require 'HTMLPurifier/DefinitionCache.php'; +require 'HTMLPurifier/DefinitionCacheFactory.php'; +require 'HTMLPurifier/Doctype.php'; +require 'HTMLPurifier/DoctypeRegistry.php'; +require 'HTMLPurifier/ElementDef.php'; +require 'HTMLPurifier/Encoder.php'; +require 'HTMLPurifier/EntityLookup.php'; +require 'HTMLPurifier/EntityParser.php'; +require 'HTMLPurifier/ErrorCollector.php'; +require 'HTMLPurifier/ErrorStruct.php'; +require 'HTMLPurifier/Exception.php'; +require 'HTMLPurifier/Filter.php'; +require 'HTMLPurifier/Generator.php'; +require 'HTMLPurifier/HTMLDefinition.php'; +require 'HTMLPurifier/HTMLModule.php'; +require 'HTMLPurifier/HTMLModuleManager.php'; +require 'HTMLPurifier/IDAccumulator.php'; +require 'HTMLPurifier/Injector.php'; +require 'HTMLPurifier/Language.php'; +require 'HTMLPurifier/LanguageFactory.php'; +require 'HTMLPurifier/Length.php'; +require 'HTMLPurifier/Lexer.php'; +require 'HTMLPurifier/PercentEncoder.php'; +require 'HTMLPurifier/PropertyList.php'; +require 'HTMLPurifier/PropertyListIterator.php'; +require 'HTMLPurifier/Strategy.php'; +require 'HTMLPurifier/StringHash.php'; +require 'HTMLPurifier/StringHashParser.php'; +require 'HTMLPurifier/TagTransform.php'; +require 'HTMLPurifier/Token.php'; +require 'HTMLPurifier/TokenFactory.php'; +require 'HTMLPurifier/URI.php'; +require 'HTMLPurifier/URIDefinition.php'; +require 'HTMLPurifier/URIFilter.php'; +require 'HTMLPurifier/URIParser.php'; +require 'HTMLPurifier/URIScheme.php'; +require 'HTMLPurifier/URISchemeRegistry.php'; +require 'HTMLPurifier/UnitConverter.php'; +require 'HTMLPurifier/VarParser.php'; +require 'HTMLPurifier/VarParserException.php'; +require 'HTMLPurifier/AttrDef/CSS.php'; +require 'HTMLPurifier/AttrDef/Clone.php'; +require 'HTMLPurifier/AttrDef/Enum.php'; +require 'HTMLPurifier/AttrDef/Integer.php'; +require 'HTMLPurifier/AttrDef/Lang.php'; +require 'HTMLPurifier/AttrDef/Switch.php'; +require 'HTMLPurifier/AttrDef/Text.php'; +require 'HTMLPurifier/AttrDef/URI.php'; +require 'HTMLPurifier/AttrDef/CSS/Number.php'; +require 'HTMLPurifier/AttrDef/CSS/AlphaValue.php'; +require 'HTMLPurifier/AttrDef/CSS/Background.php'; +require 'HTMLPurifier/AttrDef/CSS/BackgroundPosition.php'; +require 'HTMLPurifier/AttrDef/CSS/Border.php'; +require 'HTMLPurifier/AttrDef/CSS/Color.php'; +require 'HTMLPurifier/AttrDef/CSS/Composite.php'; +require 'HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php'; +require 'HTMLPurifier/AttrDef/CSS/Filter.php'; +require 'HTMLPurifier/AttrDef/CSS/Font.php'; +require 'HTMLPurifier/AttrDef/CSS/FontFamily.php'; +require 'HTMLPurifier/AttrDef/CSS/Ident.php'; +require 'HTMLPurifier/AttrDef/CSS/ImportantDecorator.php'; +require 'HTMLPurifier/AttrDef/CSS/Length.php'; +require 'HTMLPurifier/AttrDef/CSS/ListStyle.php'; +require 'HTMLPurifier/AttrDef/CSS/Multiple.php'; +require 'HTMLPurifier/AttrDef/CSS/Percentage.php'; +require 'HTMLPurifier/AttrDef/CSS/TextDecoration.php'; +require 'HTMLPurifier/AttrDef/CSS/URI.php'; +require 'HTMLPurifier/AttrDef/HTML/Bool.php'; +require 'HTMLPurifier/AttrDef/HTML/Nmtokens.php'; +require 'HTMLPurifier/AttrDef/HTML/Class.php'; +require 'HTMLPurifier/AttrDef/HTML/Color.php'; +require 'HTMLPurifier/AttrDef/HTML/FrameTarget.php'; +require 'HTMLPurifier/AttrDef/HTML/ID.php'; +require 'HTMLPurifier/AttrDef/HTML/Pixels.php'; +require 'HTMLPurifier/AttrDef/HTML/Length.php'; +require 'HTMLPurifier/AttrDef/HTML/LinkTypes.php'; +require 'HTMLPurifier/AttrDef/HTML/MultiLength.php'; +require 'HTMLPurifier/AttrDef/URI/Email.php'; +require 'HTMLPurifier/AttrDef/URI/Host.php'; +require 'HTMLPurifier/AttrDef/URI/IPv4.php'; +require 'HTMLPurifier/AttrDef/URI/IPv6.php'; +require 'HTMLPurifier/AttrDef/URI/Email/SimpleCheck.php'; +require 'HTMLPurifier/AttrTransform/Background.php'; +require 'HTMLPurifier/AttrTransform/BdoDir.php'; +require 'HTMLPurifier/AttrTransform/BgColor.php'; +require 'HTMLPurifier/AttrTransform/BoolToCSS.php'; +require 'HTMLPurifier/AttrTransform/Border.php'; +require 'HTMLPurifier/AttrTransform/EnumToCSS.php'; +require 'HTMLPurifier/AttrTransform/ImgRequired.php'; +require 'HTMLPurifier/AttrTransform/ImgSpace.php'; +require 'HTMLPurifier/AttrTransform/Input.php'; +require 'HTMLPurifier/AttrTransform/Lang.php'; +require 'HTMLPurifier/AttrTransform/Length.php'; +require 'HTMLPurifier/AttrTransform/Name.php'; +require 'HTMLPurifier/AttrTransform/NameSync.php'; +require 'HTMLPurifier/AttrTransform/Nofollow.php'; +require 'HTMLPurifier/AttrTransform/SafeEmbed.php'; +require 'HTMLPurifier/AttrTransform/SafeObject.php'; +require 'HTMLPurifier/AttrTransform/SafeParam.php'; +require 'HTMLPurifier/AttrTransform/ScriptRequired.php'; +require 'HTMLPurifier/AttrTransform/TargetBlank.php'; +require 'HTMLPurifier/AttrTransform/Textarea.php'; +require 'HTMLPurifier/ChildDef/Chameleon.php'; +require 'HTMLPurifier/ChildDef/Custom.php'; +require 'HTMLPurifier/ChildDef/Empty.php'; +require 'HTMLPurifier/ChildDef/List.php'; +require 'HTMLPurifier/ChildDef/Required.php'; +require 'HTMLPurifier/ChildDef/Optional.php'; +require 'HTMLPurifier/ChildDef/StrictBlockquote.php'; +require 'HTMLPurifier/ChildDef/Table.php'; +require 'HTMLPurifier/DefinitionCache/Decorator.php'; +require 'HTMLPurifier/DefinitionCache/Null.php'; +require 'HTMLPurifier/DefinitionCache/Serializer.php'; +require 'HTMLPurifier/DefinitionCache/Decorator/Cleanup.php'; +require 'HTMLPurifier/DefinitionCache/Decorator/Memory.php'; +require 'HTMLPurifier/HTMLModule/Bdo.php'; +require 'HTMLPurifier/HTMLModule/CommonAttributes.php'; +require 'HTMLPurifier/HTMLModule/Edit.php'; +require 'HTMLPurifier/HTMLModule/Forms.php'; +require 'HTMLPurifier/HTMLModule/Hypertext.php'; +require 'HTMLPurifier/HTMLModule/Iframe.php'; +require 'HTMLPurifier/HTMLModule/Image.php'; +require 'HTMLPurifier/HTMLModule/Legacy.php'; +require 'HTMLPurifier/HTMLModule/List.php'; +require 'HTMLPurifier/HTMLModule/Name.php'; +require 'HTMLPurifier/HTMLModule/Nofollow.php'; +require 'HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php'; +require 'HTMLPurifier/HTMLModule/Object.php'; +require 'HTMLPurifier/HTMLModule/Presentation.php'; +require 'HTMLPurifier/HTMLModule/Proprietary.php'; +require 'HTMLPurifier/HTMLModule/Ruby.php'; +require 'HTMLPurifier/HTMLModule/SafeEmbed.php'; +require 'HTMLPurifier/HTMLModule/SafeObject.php'; +require 'HTMLPurifier/HTMLModule/Scripting.php'; +require 'HTMLPurifier/HTMLModule/StyleAttribute.php'; +require 'HTMLPurifier/HTMLModule/Tables.php'; +require 'HTMLPurifier/HTMLModule/Target.php'; +require 'HTMLPurifier/HTMLModule/TargetBlank.php'; +require 'HTMLPurifier/HTMLModule/Text.php'; +require 'HTMLPurifier/HTMLModule/Tidy.php'; +require 'HTMLPurifier/HTMLModule/XMLCommonAttributes.php'; +require 'HTMLPurifier/HTMLModule/Tidy/Name.php'; +require 'HTMLPurifier/HTMLModule/Tidy/Proprietary.php'; +require 'HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php'; +require 'HTMLPurifier/HTMLModule/Tidy/Strict.php'; +require 'HTMLPurifier/HTMLModule/Tidy/Transitional.php'; +require 'HTMLPurifier/HTMLModule/Tidy/XHTML.php'; +require 'HTMLPurifier/Injector/AutoParagraph.php'; +require 'HTMLPurifier/Injector/DisplayLinkURI.php'; +require 'HTMLPurifier/Injector/Linkify.php'; +require 'HTMLPurifier/Injector/PurifierLinkify.php'; +require 'HTMLPurifier/Injector/RemoveEmpty.php'; +require 'HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php'; +require 'HTMLPurifier/Injector/SafeObject.php'; +require 'HTMLPurifier/Lexer/DOMLex.php'; +require 'HTMLPurifier/Lexer/DirectLex.php'; +require 'HTMLPurifier/Strategy/Composite.php'; +require 'HTMLPurifier/Strategy/Core.php'; +require 'HTMLPurifier/Strategy/FixNesting.php'; +require 'HTMLPurifier/Strategy/MakeWellFormed.php'; +require 'HTMLPurifier/Strategy/RemoveForeignElements.php'; +require 'HTMLPurifier/Strategy/ValidateAttributes.php'; +require 'HTMLPurifier/TagTransform/Font.php'; +require 'HTMLPurifier/TagTransform/Simple.php'; +require 'HTMLPurifier/Token/Comment.php'; +require 'HTMLPurifier/Token/Tag.php'; +require 'HTMLPurifier/Token/Empty.php'; +require 'HTMLPurifier/Token/End.php'; +require 'HTMLPurifier/Token/Start.php'; +require 'HTMLPurifier/Token/Text.php'; +require 'HTMLPurifier/URIFilter/DisableExternal.php'; +require 'HTMLPurifier/URIFilter/DisableExternalResources.php'; +require 'HTMLPurifier/URIFilter/DisableResources.php'; +require 'HTMLPurifier/URIFilter/HostBlacklist.php'; +require 'HTMLPurifier/URIFilter/MakeAbsolute.php'; +require 'HTMLPurifier/URIFilter/Munge.php'; +require 'HTMLPurifier/URIFilter/SafeIframe.php'; +require 'HTMLPurifier/URIScheme/data.php'; +require 'HTMLPurifier/URIScheme/file.php'; +require 'HTMLPurifier/URIScheme/ftp.php'; +require 'HTMLPurifier/URIScheme/http.php'; +require 'HTMLPurifier/URIScheme/https.php'; +require 'HTMLPurifier/URIScheme/mailto.php'; +require 'HTMLPurifier/URIScheme/news.php'; +require 'HTMLPurifier/URIScheme/nntp.php'; +require 'HTMLPurifier/VarParser/Flexible.php'; +require 'HTMLPurifier/VarParser/Native.php'; diff --git a/lib/htmlpurifier/library/HTMLPurifier.kses.php b/lib/htmlpurifier/library/HTMLPurifier.kses.php new file mode 100644 index 000000000..3143feb17 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.kses.php @@ -0,0 +1,30 @@ +<?php + +/** + * @file + * Emulation layer for code that used kses(), substituting in HTML Purifier. + */ + +require_once dirname(__FILE__) . '/HTMLPurifier.auto.php'; + +function kses($string, $allowed_html, $allowed_protocols = null) { + $config = HTMLPurifier_Config::createDefault(); + $allowed_elements = array(); + $allowed_attributes = array(); + foreach ($allowed_html as $element => $attributes) { + $allowed_elements[$element] = true; + foreach ($attributes as $attribute => $x) { + $allowed_attributes["$element.$attribute"] = true; + } + } + $config->set('HTML.AllowedElements', $allowed_elements); + $config->set('HTML.AllowedAttributes', $allowed_attributes); + $allowed_schemes = array(); + if ($allowed_protocols !== null) { + $config->set('URI.AllowedSchemes', $allowed_protocols); + } + $purifier = new HTMLPurifier($config); + return $purifier->purify($string); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.path.php b/lib/htmlpurifier/library/HTMLPurifier.path.php new file mode 100644 index 000000000..39b1b6531 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.path.php @@ -0,0 +1,11 @@ +<?php + +/** + * @file + * Convenience stub file that adds HTML Purifier's library file to the path + * without any other side-effects. + */ + +set_include_path(dirname(__FILE__) . PATH_SEPARATOR . get_include_path() ); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.php b/lib/htmlpurifier/library/HTMLPurifier.php new file mode 100644 index 000000000..e599e1c0c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.php @@ -0,0 +1,237 @@ +<?php + +/*! @mainpage + * + * HTML Purifier is an HTML filter that will take an arbitrary snippet of + * HTML and rigorously test, validate and filter it into a version that + * is safe for output onto webpages. It achieves this by: + * + * -# Lexing (parsing into tokens) the document, + * -# Executing various strategies on the tokens: + * -# Removing all elements not in the whitelist, + * -# Making the tokens well-formed, + * -# Fixing the nesting of the nodes, and + * -# Validating attributes of the nodes; and + * -# Generating HTML from the purified tokens. + * + * However, most users will only need to interface with the HTMLPurifier + * and HTMLPurifier_Config. + */ + +/* + HTML Purifier 4.4.0 - Standards Compliant HTML Filtering + Copyright (C) 2006-2008 Edward Z. Yang + + This library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with this library; if not, write to the Free Software + Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +/** + * Facade that coordinates HTML Purifier's subsystems in order to purify HTML. + * + * @note There are several points in which configuration can be specified + * for HTML Purifier. The precedence of these (from lowest to + * highest) is as follows: + * -# Instance: new HTMLPurifier($config) + * -# Invocation: purify($html, $config) + * These configurations are entirely independent of each other and + * are *not* merged (this behavior may change in the future). + * + * @todo We need an easier way to inject strategies using the configuration + * object. + */ +class HTMLPurifier +{ + + /** Version of HTML Purifier */ + public $version = '4.4.0'; + + /** Constant with version of HTML Purifier */ + const VERSION = '4.4.0'; + + /** Global configuration object */ + public $config; + + /** Array of extra HTMLPurifier_Filter objects to run on HTML, for backwards compatibility */ + private $filters = array(); + + /** Single instance of HTML Purifier */ + private static $instance; + + protected $strategy, $generator; + + /** + * Resultant HTMLPurifier_Context of last run purification. Is an array + * of contexts if the last called method was purifyArray(). + */ + public $context; + + /** + * Initializes the purifier. + * @param $config Optional HTMLPurifier_Config object for all instances of + * the purifier, if omitted, a default configuration is + * supplied (which can be overridden on a per-use basis). + * The parameter can also be any type that + * HTMLPurifier_Config::create() supports. + */ + public function __construct($config = null) { + + $this->config = HTMLPurifier_Config::create($config); + + $this->strategy = new HTMLPurifier_Strategy_Core(); + + } + + /** + * Adds a filter to process the output. First come first serve + * @param $filter HTMLPurifier_Filter object + */ + public function addFilter($filter) { + trigger_error('HTMLPurifier->addFilter() is deprecated, use configuration directives in the Filter namespace or Filter.Custom', E_USER_WARNING); + $this->filters[] = $filter; + } + + /** + * Filters an HTML snippet/document to be XSS-free and standards-compliant. + * + * @param $html String of HTML to purify + * @param $config HTMLPurifier_Config object for this operation, if omitted, + * defaults to the config object specified during this + * object's construction. The parameter can also be any type + * that HTMLPurifier_Config::create() supports. + * @return Purified HTML + */ + public function purify($html, $config = null) { + + // :TODO: make the config merge in, instead of replace + $config = $config ? HTMLPurifier_Config::create($config) : $this->config; + + // implementation is partially environment dependant, partially + // configuration dependant + $lexer = HTMLPurifier_Lexer::create($config); + + $context = new HTMLPurifier_Context(); + + // setup HTML generator + $this->generator = new HTMLPurifier_Generator($config, $context); + $context->register('Generator', $this->generator); + + // set up global context variables + if ($config->get('Core.CollectErrors')) { + // may get moved out if other facilities use it + $language_factory = HTMLPurifier_LanguageFactory::instance(); + $language = $language_factory->create($config, $context); + $context->register('Locale', $language); + + $error_collector = new HTMLPurifier_ErrorCollector($context); + $context->register('ErrorCollector', $error_collector); + } + + // setup id_accumulator context, necessary due to the fact that + // AttrValidator can be called from many places + $id_accumulator = HTMLPurifier_IDAccumulator::build($config, $context); + $context->register('IDAccumulator', $id_accumulator); + + $html = HTMLPurifier_Encoder::convertToUTF8($html, $config, $context); + + // setup filters + $filter_flags = $config->getBatch('Filter'); + $custom_filters = $filter_flags['Custom']; + unset($filter_flags['Custom']); + $filters = array(); + foreach ($filter_flags as $filter => $flag) { + if (!$flag) continue; + if (strpos($filter, '.') !== false) continue; + $class = "HTMLPurifier_Filter_$filter"; + $filters[] = new $class; + } + foreach ($custom_filters as $filter) { + // maybe "HTMLPurifier_Filter_$filter", but be consistent with AutoFormat + $filters[] = $filter; + } + $filters = array_merge($filters, $this->filters); + // maybe prepare(), but later + + for ($i = 0, $filter_size = count($filters); $i < $filter_size; $i++) { + $html = $filters[$i]->preFilter($html, $config, $context); + } + + // purified HTML + $html = + $this->generator->generateFromTokens( + // list of tokens + $this->strategy->execute( + // list of un-purified tokens + $lexer->tokenizeHTML( + // un-purified HTML + $html, $config, $context + ), + $config, $context + ) + ); + + for ($i = $filter_size - 1; $i >= 0; $i--) { + $html = $filters[$i]->postFilter($html, $config, $context); + } + + $html = HTMLPurifier_Encoder::convertFromUTF8($html, $config, $context); + $this->context =& $context; + return $html; + } + + /** + * Filters an array of HTML snippets + * @param $config Optional HTMLPurifier_Config object for this operation. + * See HTMLPurifier::purify() for more details. + * @return Array of purified HTML + */ + public function purifyArray($array_of_html, $config = null) { + $context_array = array(); + foreach ($array_of_html as $key => $html) { + $array_of_html[$key] = $this->purify($html, $config); + $context_array[$key] = $this->context; + } + $this->context = $context_array; + return $array_of_html; + } + + /** + * Singleton for enforcing just one HTML Purifier in your system + * @param $prototype Optional prototype HTMLPurifier instance to + * overload singleton with, or HTMLPurifier_Config + * instance to configure the generated version with. + */ + public static function instance($prototype = null) { + if (!self::$instance || $prototype) { + if ($prototype instanceof HTMLPurifier) { + self::$instance = $prototype; + } elseif ($prototype) { + self::$instance = new HTMLPurifier($prototype); + } else { + self::$instance = new HTMLPurifier(); + } + } + return self::$instance; + } + + /** + * @note Backwards compatibility, see instance() + */ + public static function getInstance($prototype = null) { + return HTMLPurifier::instance($prototype); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier.safe-includes.php b/lib/htmlpurifier/library/HTMLPurifier.safe-includes.php new file mode 100644 index 000000000..d49b196c4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier.safe-includes.php @@ -0,0 +1,215 @@ +<?php + +/** + * @file + * This file was auto-generated by generate-includes.php and includes all of + * the core files required by HTML Purifier. This is a convenience stub that + * includes all files using dirname(__FILE__) and require_once. PLEASE DO NOT + * EDIT THIS FILE, changes will be overwritten the next time the script is run. + * + * Changes to include_path are not necessary. + */ + +$__dir = dirname(__FILE__); + +require_once $__dir . '/HTMLPurifier.php'; +require_once $__dir . '/HTMLPurifier/AttrCollections.php'; +require_once $__dir . '/HTMLPurifier/AttrDef.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform.php'; +require_once $__dir . '/HTMLPurifier/AttrTypes.php'; +require_once $__dir . '/HTMLPurifier/AttrValidator.php'; +require_once $__dir . '/HTMLPurifier/Bootstrap.php'; +require_once $__dir . '/HTMLPurifier/Definition.php'; +require_once $__dir . '/HTMLPurifier/CSSDefinition.php'; +require_once $__dir . '/HTMLPurifier/ChildDef.php'; +require_once $__dir . '/HTMLPurifier/Config.php'; +require_once $__dir . '/HTMLPurifier/ConfigSchema.php'; +require_once $__dir . '/HTMLPurifier/ContentSets.php'; +require_once $__dir . '/HTMLPurifier/Context.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCache.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCacheFactory.php'; +require_once $__dir . '/HTMLPurifier/Doctype.php'; +require_once $__dir . '/HTMLPurifier/DoctypeRegistry.php'; +require_once $__dir . '/HTMLPurifier/ElementDef.php'; +require_once $__dir . '/HTMLPurifier/Encoder.php'; +require_once $__dir . '/HTMLPurifier/EntityLookup.php'; +require_once $__dir . '/HTMLPurifier/EntityParser.php'; +require_once $__dir . '/HTMLPurifier/ErrorCollector.php'; +require_once $__dir . '/HTMLPurifier/ErrorStruct.php'; +require_once $__dir . '/HTMLPurifier/Exception.php'; +require_once $__dir . '/HTMLPurifier/Filter.php'; +require_once $__dir . '/HTMLPurifier/Generator.php'; +require_once $__dir . '/HTMLPurifier/HTMLDefinition.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule.php'; +require_once $__dir . '/HTMLPurifier/HTMLModuleManager.php'; +require_once $__dir . '/HTMLPurifier/IDAccumulator.php'; +require_once $__dir . '/HTMLPurifier/Injector.php'; +require_once $__dir . '/HTMLPurifier/Language.php'; +require_once $__dir . '/HTMLPurifier/LanguageFactory.php'; +require_once $__dir . '/HTMLPurifier/Length.php'; +require_once $__dir . '/HTMLPurifier/Lexer.php'; +require_once $__dir . '/HTMLPurifier/PercentEncoder.php'; +require_once $__dir . '/HTMLPurifier/PropertyList.php'; +require_once $__dir . '/HTMLPurifier/PropertyListIterator.php'; +require_once $__dir . '/HTMLPurifier/Strategy.php'; +require_once $__dir . '/HTMLPurifier/StringHash.php'; +require_once $__dir . '/HTMLPurifier/StringHashParser.php'; +require_once $__dir . '/HTMLPurifier/TagTransform.php'; +require_once $__dir . '/HTMLPurifier/Token.php'; +require_once $__dir . '/HTMLPurifier/TokenFactory.php'; +require_once $__dir . '/HTMLPurifier/URI.php'; +require_once $__dir . '/HTMLPurifier/URIDefinition.php'; +require_once $__dir . '/HTMLPurifier/URIFilter.php'; +require_once $__dir . '/HTMLPurifier/URIParser.php'; +require_once $__dir . '/HTMLPurifier/URIScheme.php'; +require_once $__dir . '/HTMLPurifier/URISchemeRegistry.php'; +require_once $__dir . '/HTMLPurifier/UnitConverter.php'; +require_once $__dir . '/HTMLPurifier/VarParser.php'; +require_once $__dir . '/HTMLPurifier/VarParserException.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/Clone.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/Enum.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/Integer.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/Lang.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/Switch.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/Text.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/URI.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Number.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/AlphaValue.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Background.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Border.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Color.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Composite.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Filter.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Font.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/FontFamily.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Ident.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Length.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/ListStyle.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Multiple.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/Percentage.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/TextDecoration.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/CSS/URI.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/Bool.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/Nmtokens.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/Class.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/Color.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/FrameTarget.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/ID.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/Pixels.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/Length.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/LinkTypes.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/HTML/MultiLength.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/URI/Email.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/URI/Host.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/URI/IPv4.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/URI/IPv6.php'; +require_once $__dir . '/HTMLPurifier/AttrDef/URI/Email/SimpleCheck.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Background.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/BdoDir.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/BgColor.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/BoolToCSS.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Border.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/EnumToCSS.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/ImgRequired.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/ImgSpace.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Input.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Lang.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Length.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Name.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/NameSync.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/Nofollow.php'; +require_once $__dir . '/HTMLPurifier/AttrTransform/SafeEmbed.php'; +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/Textarea.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/Chameleon.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/Custom.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/Empty.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/List.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/Required.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/Optional.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/StrictBlockquote.php'; +require_once $__dir . '/HTMLPurifier/ChildDef/Table.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCache/Decorator.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCache/Null.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCache/Serializer.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCache/Decorator/Cleanup.php'; +require_once $__dir . '/HTMLPurifier/DefinitionCache/Decorator/Memory.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Bdo.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/CommonAttributes.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Edit.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Forms.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Hypertext.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Iframe.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Image.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Legacy.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/List.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Name.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Nofollow.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Object.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Presentation.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Proprietary.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Ruby.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/SafeEmbed.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/SafeObject.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Scripting.php'; +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/Text.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/XMLCommonAttributes.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy/Name.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy/Proprietary.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy/Strict.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy/Transitional.php'; +require_once $__dir . '/HTMLPurifier/HTMLModule/Tidy/XHTML.php'; +require_once $__dir . '/HTMLPurifier/Injector/AutoParagraph.php'; +require_once $__dir . '/HTMLPurifier/Injector/DisplayLinkURI.php'; +require_once $__dir . '/HTMLPurifier/Injector/Linkify.php'; +require_once $__dir . '/HTMLPurifier/Injector/PurifierLinkify.php'; +require_once $__dir . '/HTMLPurifier/Injector/RemoveEmpty.php'; +require_once $__dir . '/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php'; +require_once $__dir . '/HTMLPurifier/Injector/SafeObject.php'; +require_once $__dir . '/HTMLPurifier/Lexer/DOMLex.php'; +require_once $__dir . '/HTMLPurifier/Lexer/DirectLex.php'; +require_once $__dir . '/HTMLPurifier/Strategy/Composite.php'; +require_once $__dir . '/HTMLPurifier/Strategy/Core.php'; +require_once $__dir . '/HTMLPurifier/Strategy/FixNesting.php'; +require_once $__dir . '/HTMLPurifier/Strategy/MakeWellFormed.php'; +require_once $__dir . '/HTMLPurifier/Strategy/RemoveForeignElements.php'; +require_once $__dir . '/HTMLPurifier/Strategy/ValidateAttributes.php'; +require_once $__dir . '/HTMLPurifier/TagTransform/Font.php'; +require_once $__dir . '/HTMLPurifier/TagTransform/Simple.php'; +require_once $__dir . '/HTMLPurifier/Token/Comment.php'; +require_once $__dir . '/HTMLPurifier/Token/Tag.php'; +require_once $__dir . '/HTMLPurifier/Token/Empty.php'; +require_once $__dir . '/HTMLPurifier/Token/End.php'; +require_once $__dir . '/HTMLPurifier/Token/Start.php'; +require_once $__dir . '/HTMLPurifier/Token/Text.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/DisableExternal.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/DisableExternalResources.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/DisableResources.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/HostBlacklist.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/MakeAbsolute.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/Munge.php'; +require_once $__dir . '/HTMLPurifier/URIFilter/SafeIframe.php'; +require_once $__dir . '/HTMLPurifier/URIScheme/data.php'; +require_once $__dir . '/HTMLPurifier/URIScheme/file.php'; +require_once $__dir . '/HTMLPurifier/URIScheme/ftp.php'; +require_once $__dir . '/HTMLPurifier/URIScheme/http.php'; +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/VarParser/Flexible.php'; +require_once $__dir . '/HTMLPurifier/VarParser/Native.php'; diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrCollections.php b/lib/htmlpurifier/library/HTMLPurifier/AttrCollections.php new file mode 100644 index 000000000..555b86d04 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrCollections.php @@ -0,0 +1,128 @@ +<?php + +/** + * Defines common attribute collections that modules reference + */ + +class HTMLPurifier_AttrCollections +{ + + /** + * Associative array of attribute collections, indexed by name + */ + public $info = array(); + + /** + * Performs all expansions on internal data for use by other inclusions + * It also collects all attribute collection extensions from + * modules + * @param $attr_types HTMLPurifier_AttrTypes instance + * @param $modules Hash array of HTMLPurifier_HTMLModule members + */ + public function __construct($attr_types, $modules) { + // load extensions from the modules + foreach ($modules as $module) { + foreach ($module->attr_collections as $coll_i => $coll) { + if (!isset($this->info[$coll_i])) { + $this->info[$coll_i] = array(); + } + foreach ($coll as $attr_i => $attr) { + if ($attr_i === 0 && isset($this->info[$coll_i][$attr_i])) { + // merge in includes + $this->info[$coll_i][$attr_i] = array_merge( + $this->info[$coll_i][$attr_i], $attr); + continue; + } + $this->info[$coll_i][$attr_i] = $attr; + } + } + } + // perform internal expansions and inclusions + foreach ($this->info as $name => $attr) { + // merge attribute collections that include others + $this->performInclusions($this->info[$name]); + // replace string identifiers with actual attribute objects + $this->expandIdentifiers($this->info[$name], $attr_types); + } + } + + /** + * Takes a reference to an attribute associative array and performs + * all inclusions specified by the zero index. + * @param &$attr Reference to attribute array + */ + public function performInclusions(&$attr) { + if (!isset($attr[0])) return; + $merge = $attr[0]; + $seen = array(); // recursion guard + // loop through all the inclusions + for ($i = 0; isset($merge[$i]); $i++) { + if (isset($seen[$merge[$i]])) continue; + $seen[$merge[$i]] = true; + // foreach attribute of the inclusion, copy it over + if (!isset($this->info[$merge[$i]])) continue; + foreach ($this->info[$merge[$i]] as $key => $value) { + if (isset($attr[$key])) continue; // also catches more inclusions + $attr[$key] = $value; + } + if (isset($this->info[$merge[$i]][0])) { + // recursion + $merge = array_merge($merge, $this->info[$merge[$i]][0]); + } + } + unset($attr[0]); + } + + /** + * Expands all string identifiers in an attribute array by replacing + * them with the appropriate values inside HTMLPurifier_AttrTypes + * @param &$attr Reference to attribute array + * @param $attr_types HTMLPurifier_AttrTypes instance + */ + public function expandIdentifiers(&$attr, $attr_types) { + + // because foreach will process new elements we add, make sure we + // skip duplicates + $processed = array(); + + foreach ($attr as $def_i => $def) { + // skip inclusions + if ($def_i === 0) continue; + + if (isset($processed[$def_i])) continue; + + // determine whether or not attribute is required + if ($required = (strpos($def_i, '*') !== false)) { + // rename the definition + unset($attr[$def_i]); + $def_i = trim($def_i, '*'); + $attr[$def_i] = $def; + } + + $processed[$def_i] = true; + + // if we've already got a literal object, move on + if (is_object($def)) { + // preserve previous required + $attr[$def_i]->required = ($required || $attr[$def_i]->required); + continue; + } + + if ($def === false) { + unset($attr[$def_i]); + continue; + } + + if ($t = $attr_types->get($def)) { + $attr[$def_i] = $t; + $attr[$def_i]->required = $required; + } else { + unset($attr[$def_i]); + } + } + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef.php new file mode 100644 index 000000000..b2e4f36c5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef.php @@ -0,0 +1,123 @@ +<?php + +/** + * Base class for all validating attribute definitions. + * + * This family of classes forms the core for not only HTML attribute validation, + * but also any sort of string that needs to be validated or cleaned (which + * means CSS properties and composite definitions are defined here too). + * Besides defining (through code) what precisely makes the string valid, + * subclasses are also responsible for cleaning the code if possible. + */ + +abstract class HTMLPurifier_AttrDef +{ + + /** + * Tells us whether or not an HTML attribute is minimized. Has no + * meaning in other contexts. + */ + public $minimized = false; + + /** + * Tells us whether or not an HTML attribute is required. Has no + * meaning in other contexts + */ + public $required = false; + + /** + * Validates and cleans passed string according to a definition. + * + * @param $string String to be validated and cleaned. + * @param $config Mandatory HTMLPurifier_Config object. + * @param $context Mandatory HTMLPurifier_AttrContext object. + */ + abstract public function validate($string, $config, $context); + + /** + * Convenience method that parses a string as if it were CDATA. + * + * This method process a string in the manner specified at + * <http://www.w3.org/TR/html4/types.html#h-6.2> by removing + * leading and trailing whitespace, ignoring line feeds, and replacing + * carriage returns and tabs with spaces. While most useful for HTML + * attributes specified as CDATA, it can also be applied to most CSS + * values. + * + * @note This method is not entirely standards compliant, as trim() removes + * more types of whitespace than specified in the spec. In practice, + * this is rarely a problem, as those extra characters usually have + * already been removed by HTMLPurifier_Encoder. + * + * @warning This processing is inconsistent with XML's whitespace handling + * as specified by section 3.3.3 and referenced XHTML 1.0 section + * 4.7. However, note that we are NOT necessarily + * parsing XML, thus, this behavior may still be correct. We + * assume that newlines have been normalized. + */ + public function parseCDATA($string) { + $string = trim($string); + $string = str_replace(array("\n", "\t", "\r"), ' ', $string); + return $string; + } + + /** + * Factory method for creating this class from a string. + * @param $string String construction info + * @return Created AttrDef object corresponding to $string + */ + public function make($string) { + // default implementation, return a flyweight of this object. + // If $string has an effect on the returned object (i.e. you + // need to overload this method), it is best + // to clone or instantiate new copies. (Instantiation is safer.) + return $this; + } + + /** + * Removes spaces from rgb(0, 0, 0) so that shorthand CSS properties work + * properly. THIS IS A HACK! + */ + protected function mungeRgb($string) { + return preg_replace('/rgb\((\d+)\s*,\s*(\d+)\s*,\s*(\d+)\)/', 'rgb(\1,\2,\3)', $string); + } + + /** + * Parses a possibly escaped CSS string and returns the "pure" + * version of it. + */ + protected function expandCSSEscape($string) { + // flexibly parse it + $ret = ''; + for ($i = 0, $c = strlen($string); $i < $c; $i++) { + if ($string[$i] === '\\') { + $i++; + if ($i >= $c) { + $ret .= '\\'; + break; + } + if (ctype_xdigit($string[$i])) { + $code = $string[$i]; + for ($a = 1, $i++; $i < $c && $a < 6; $i++, $a++) { + if (!ctype_xdigit($string[$i])) break; + $code .= $string[$i]; + } + // We have to be extremely careful when adding + // new characters, to make sure we're not breaking + // the encoding. + $char = HTMLPurifier_Encoder::unichr(hexdec($code)); + if (HTMLPurifier_Encoder::cleanUTF8($char) === '') continue; + $ret .= $char; + if ($i < $c && trim($string[$i]) !== '') $i--; + continue; + } + if ($string[$i] === "\n") continue; + } + $ret .= $string[$i]; + } + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS.php new file mode 100644 index 000000000..953e70675 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS.php @@ -0,0 +1,87 @@ +<?php + +/** + * Validates the HTML attribute style, otherwise known as CSS. + * @note We don't implement the whole CSS specification, so it might be + * difficult to reuse this component in the context of validating + * actual stylesheet declarations. + * @note If we were really serious about validating the CSS, we would + * tokenize the styles and then parse the tokens. Obviously, we + * are not doing that. Doing that could seriously harm performance, + * but would make these components a lot more viable for a CSS + * filtering solution. + */ +class HTMLPurifier_AttrDef_CSS extends HTMLPurifier_AttrDef +{ + + public function validate($css, $config, $context) { + + $css = $this->parseCDATA($css); + + $definition = $config->getCSSDefinition(); + + // 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); + $propvalues = array(); + + /** + * Name of the current CSS property being validated. + */ + $property = false; + $context->register('CurrentCSSProperty', $property); + + foreach ($declarations as $declaration) { + if (!$declaration) continue; + if (!strpos($declaration, ':')) continue; + list($property, $value) = explode(':', $declaration, 2); + $property = trim($property); + $value = trim($value); + $ok = false; + do { + if (isset($definition->info[$property])) { + $ok = true; + break; + } + if (ctype_lower($property)) break; + $property = strtolower($property); + if (isset($definition->info[$property])) { + $ok = true; + break; + } + } while(0); + if (!$ok) continue; + // inefficient call, since the validator will do this again + if (strtolower(trim($value)) !== 'inherit') { + // inherit works for everything (but only on the base property) + $result = $definition->info[$property]->validate( + $value, $config, $context ); + } else { + $result = 'inherit'; + } + if ($result === false) continue; + $propvalues[$property] = $result; + } + + $context->destroy('CurrentCSSProperty'); + + // procedure does not write the new CSS simultaneously, so it's + // 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;"; + } + + return $new_declarations ? $new_declarations : false; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php new file mode 100644 index 000000000..292c040d4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/AlphaValue.php @@ -0,0 +1,21 @@ +<?php + +class HTMLPurifier_AttrDef_CSS_AlphaValue extends HTMLPurifier_AttrDef_CSS_Number +{ + + public function __construct() { + parent::__construct(false); // opacity is non-negative, but we will clamp it + } + + public function validate($number, $config, $context) { + $result = parent::validate($number, $config, $context); + if ($result === false) return $result; + $float = (float) $result; + if ($float < 0.0) $result = '0'; + if ($float > 1.0) $result = '1'; + return $result; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php new file mode 100644 index 000000000..3a3d20cd6 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Background.php @@ -0,0 +1,87 @@ +<?php + +/** + * Validates shorthand CSS property background. + * @warning Does not support url tokens that have internal spaces. + */ +class HTMLPurifier_AttrDef_CSS_Background extends HTMLPurifier_AttrDef +{ + + /** + * Local copy of component validators. + * @note See HTMLPurifier_AttrDef_Font::$info for a similar impl. + */ + protected $info; + + public function __construct($config) { + $def = $config->getCSSDefinition(); + $this->info['background-color'] = $def->info['background-color']; + $this->info['background-image'] = $def->info['background-image']; + $this->info['background-repeat'] = $def->info['background-repeat']; + $this->info['background-attachment'] = $def->info['background-attachment']; + $this->info['background-position'] = $def->info['background-position']; + } + + public function validate($string, $config, $context) { + + // regular pre-processing + $string = $this->parseCDATA($string); + if ($string === '') return false; + + // munge rgb() decl if necessary + $string = $this->mungeRgb($string); + + // assumes URI doesn't have spaces in it + $bits = explode(' ', strtolower($string)); // bits to process + + $caught = array(); + $caught['color'] = false; + $caught['image'] = false; + $caught['repeat'] = false; + $caught['attachment'] = false; + $caught['position'] = false; + + $i = 0; // number of catches + $none = false; + + foreach ($bits as $bit) { + if ($bit === '') continue; + foreach ($caught as $key => $status) { + if ($key != 'position') { + if ($status !== false) continue; + $r = $this->info['background-' . $key]->validate($bit, $config, $context); + } else { + $r = $bit; + } + if ($r === false) continue; + if ($key == 'position') { + if ($caught[$key] === false) $caught[$key] = ''; + $caught[$key] .= $r . ' '; + } else { + $caught[$key] = $r; + } + $i++; + break; + } + } + + if (!$i) return false; + if ($caught['position'] !== false) { + $caught['position'] = $this->info['background-position']-> + validate($caught['position'], $config, $context); + } + + $ret = array(); + foreach ($caught as $value) { + if ($value === false) continue; + $ret[] = $value; + } + + if (empty($ret)) return false; + return implode(' ', $ret); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php new file mode 100644 index 000000000..fae82eaec --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/BackgroundPosition.php @@ -0,0 +1,133 @@ +<?php + +/* W3C says: + [ // adjective and number must be in correct order, even if + // you could switch them without introducing ambiguity. + // some browsers support that syntax + [ + <percentage> | <length> | left | center | right + ] + [ + <percentage> | <length> | top | center | bottom + ]? + ] | + [ // this signifies that the vertical and horizontal adjectives + // can be arbitrarily ordered, however, there can only be two, + // one of each, or none at all + [ + left | center | right + ] || + [ + top | center | bottom + ] + ] + top, left = 0% + center, (none) = 50% + bottom, right = 100% +*/ + +/* QuirksMode says: + keyword + length/percentage must be ordered correctly, as per W3C + + Internet Explorer and Opera, however, support arbitrary ordering. We + should fix it up. + + Minor issue though, not strictly necessary. +*/ + +// control freaks may appreciate the ability to convert these to +// percentages or something, but it's not necessary + +/** + * Validates the value of background-position. + */ +class HTMLPurifier_AttrDef_CSS_BackgroundPosition extends HTMLPurifier_AttrDef +{ + + protected $length; + protected $percentage; + + public function __construct() { + $this->length = new HTMLPurifier_AttrDef_CSS_Length(); + $this->percentage = new HTMLPurifier_AttrDef_CSS_Percentage(); + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + $bits = explode(' ', $string); + + $keywords = array(); + $keywords['h'] = false; // left, right + $keywords['v'] = false; // top, bottom + $keywords['ch'] = false; // center (first word) + $keywords['cv'] = false; // center (second word) + $measures = array(); + + $i = 0; + + $lookup = array( + 'top' => 'v', + 'bottom' => 'v', + 'left' => 'h', + 'right' => 'h', + 'center' => 'c' + ); + + foreach ($bits as $bit) { + if ($bit === '') continue; + + // test for keyword + $lbit = ctype_lower($bit) ? $bit : strtolower($bit); + if (isset($lookup[$lbit])) { + $status = $lookup[$lbit]; + if ($status == 'c') { + if ($i == 0) { + $status = 'ch'; + } else { + $status = 'cv'; + } + } + $keywords[$status] = $lbit; + $i++; + } + + // test for length + $r = $this->length->validate($bit, $config, $context); + if ($r !== false) { + $measures[] = $r; + $i++; + } + + // test for percentage + $r = $this->percentage->validate($bit, $config, $context); + if ($r !== false) { + $measures[] = $r; + $i++; + } + + } + + if (!$i) return false; // no valid values were caught + + $ret = array(); + + // first keyword + if ($keywords['h']) $ret[] = $keywords['h']; + elseif ($keywords['ch']) { + $ret[] = $keywords['ch']; + $keywords['cv'] = false; // prevent re-use: center = center center + } + elseif (count($measures)) $ret[] = array_shift($measures); + + if ($keywords['v']) $ret[] = $keywords['v']; + elseif ($keywords['cv']) $ret[] = $keywords['cv']; + elseif (count($measures)) $ret[] = array_shift($measures); + + if (empty($ret)) return false; + return implode(' ', $ret); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php new file mode 100644 index 000000000..42a1d1b4a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Border.php @@ -0,0 +1,43 @@ +<?php + +/** + * Validates the border property as defined by CSS. + */ +class HTMLPurifier_AttrDef_CSS_Border extends HTMLPurifier_AttrDef +{ + + /** + * Local copy of properties this property is shorthand for. + */ + protected $info = array(); + + public function __construct($config) { + $def = $config->getCSSDefinition(); + $this->info['border-width'] = $def->info['border-width']; + $this->info['border-style'] = $def->info['border-style']; + $this->info['border-top-color'] = $def->info['border-top-color']; + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + $string = $this->mungeRgb($string); + $bits = explode(' ', $string); + $done = array(); // segments we've finished + $ret = ''; // return value + foreach ($bits as $bit) { + foreach ($this->info as $propname => $validator) { + if (isset($done[$propname])) continue; + $r = $validator->validate($bit, $config, $context); + if ($r !== false) { + $ret .= $r . ' '; + $done[$propname] = true; + break; + } + } + } + return rtrim($ret); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php new file mode 100644 index 000000000..07f95a671 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Color.php @@ -0,0 +1,78 @@ +<?php + +/** + * Validates Color as defined by CSS. + */ +class HTMLPurifier_AttrDef_CSS_Color extends HTMLPurifier_AttrDef +{ + + public function validate($color, $config, $context) { + + static $colors = null; + if ($colors === null) $colors = $config->get('Core.ColorKeywords'); + + $color = trim($color); + if ($color === '') return false; + + $lower = strtolower($color); + if (isset($colors[$lower])) return $colors[$lower]; + + if (strpos($color, 'rgb(') !== false) { + // rgb literal handling + $length = strlen($color); + if (strpos($color, ')') !== $length - 1) return false; + $triad = substr($color, 4, $length - 4 - 1); + $parts = explode(',', $triad); + if (count($parts) !== 3) return false; + $type = false; // to ensure that they're all the same type + $new_parts = array(); + foreach ($parts as $part) { + $part = trim($part); + if ($part === '') return false; + $length = strlen($part); + if ($part[$length - 1] === '%') { + // handle percents + if (!$type) { + $type = 'percentage'; + } elseif ($type !== 'percentage') { + return false; + } + $num = (float) substr($part, 0, $length - 1); + if ($num < 0) $num = 0; + if ($num > 100) $num = 100; + $new_parts[] = "$num%"; + } 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; + } + } + $new_triad = implode(',', $new_parts); + $color = "rgb($new_triad)"; + } else { + // hexadecimal handling + if ($color[0] === '#') { + $hex = substr($color, 1); + } else { + $hex = $color; + $color = '#' . $color; + } + $length = strlen($hex); + if ($length !== 3 && $length !== 6) return false; + if (!ctype_xdigit($hex)) return false; + } + + return $color; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php new file mode 100644 index 000000000..de1289cba --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Composite.php @@ -0,0 +1,38 @@ +<?php + +/** + * Allows multiple validators to attempt to validate attribute. + * + * Composite is just what it sounds like: a composite of many validators. + * This means that multiple HTMLPurifier_AttrDef objects will have a whack + * at the string. If one of them passes, that's what is returned. This is + * especially useful for CSS values, which often are a choice between + * an enumerated set of predefined values or a flexible data type. + */ +class HTMLPurifier_AttrDef_CSS_Composite extends HTMLPurifier_AttrDef +{ + + /** + * List of HTMLPurifier_AttrDef objects that may process strings + * @todo Make protected + */ + public $defs; + + /** + * @param $defs List of HTMLPurifier_AttrDef objects + */ + public function __construct($defs) { + $this->defs = $defs; + } + + public function validate($string, $config, $context) { + foreach ($this->defs as $i => $def) { + $result = $this->defs[$i]->validate($string, $config, $context); + if ($result !== false) return $result; + } + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php new file mode 100644 index 000000000..6599c5b2d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/DenyElementDecorator.php @@ -0,0 +1,28 @@ +<?php + +/** + * Decorator which enables CSS properties to be disabled for specific elements. + */ +class HTMLPurifier_AttrDef_CSS_DenyElementDecorator extends HTMLPurifier_AttrDef +{ + public $def, $element; + + /** + * @param $def Definition to wrap + * @param $element Element to deny + */ + public function __construct($def, $element) { + $this->def = $def; + $this->element = $element; + } + /** + * Checks if CurrentToken is set and equal to $this->element + */ + public function validate($string, $config, $context) { + $token = $context->get('CurrentToken', true); + if ($token && $token->name == $this->element) return false; + return $this->def->validate($string, $config, $context); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php new file mode 100644 index 000000000..147894b86 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Filter.php @@ -0,0 +1,54 @@ +<?php + +/** + * Microsoft's proprietary filter: CSS property + * @note Currently supports the alpha filter. In the future, this will + * probably need an extensible framework + */ +class HTMLPurifier_AttrDef_CSS_Filter extends HTMLPurifier_AttrDef +{ + + protected $intValidator; + + public function __construct() { + $this->intValidator = new HTMLPurifier_AttrDef_Integer(); + } + + public function validate($value, $config, $context) { + $value = $this->parseCDATA($value); + if ($value === 'none') return $value; + // if we looped this we could support multiple filters + $function_length = strcspn($value, '('); + $function = trim(substr($value, 0, $function_length)); + if ($function !== 'alpha' && + $function !== 'Alpha' && + $function !== 'progid:DXImageTransform.Microsoft.Alpha' + ) return false; + $cursor = $function_length + 1; + $parameters_length = strcspn($value, ')', $cursor); + $parameters = substr($value, $cursor, $parameters_length); + $params = explode(',', $parameters); + $ret_params = array(); + $lookup = array(); + foreach ($params as $param) { + list($key, $value) = explode('=', $param); + $key = trim($key); + $value = trim($value); + if (isset($lookup[$key])) continue; + if ($key !== 'opacity') continue; + $value = $this->intValidator->validate($value, $config, $context); + if ($value === false) continue; + $int = (int) $value; + if ($int > 100) $value = '100'; + if ($int < 0) $value = '0'; + $ret_params[] = "$key=$value"; + $lookup[$key] = true; + } + $ret_parameters = implode(',', $ret_params); + $ret_function = "$function($ret_parameters)"; + return $ret_function; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php new file mode 100644 index 000000000..699ee0b70 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Font.php @@ -0,0 +1,149 @@ +<?php + +/** + * Validates shorthand CSS property font. + */ +class HTMLPurifier_AttrDef_CSS_Font extends HTMLPurifier_AttrDef +{ + + /** + * Local copy of component validators. + * + * @note If we moved specific CSS property definitions to their own + * classes instead of having them be assembled at run time by + * CSSDefinition, this wouldn't be necessary. We'd instantiate + * our own copies. + */ + protected $info = array(); + + public function __construct($config) { + $def = $config->getCSSDefinition(); + $this->info['font-style'] = $def->info['font-style']; + $this->info['font-variant'] = $def->info['font-variant']; + $this->info['font-weight'] = $def->info['font-weight']; + $this->info['font-size'] = $def->info['font-size']; + $this->info['line-height'] = $def->info['line-height']; + $this->info['font-family'] = $def->info['font-family']; + } + + public function validate($string, $config, $context) { + + static $system_fonts = array( + 'caption' => true, + 'icon' => true, + 'menu' => true, + 'message-box' => true, + 'small-caption' => true, + 'status-bar' => true + ); + + // regular pre-processing + $string = $this->parseCDATA($string); + if ($string === '') return false; + + // check if it's one of the keywords + $lowercase_string = strtolower($string); + if (isset($system_fonts[$lowercase_string])) { + return $lowercase_string; + } + + $bits = explode(' ', $string); // bits to process + $stage = 0; // this indicates what we're looking for + $caught = array(); // which stage 0 properties have we caught? + $stage_1 = array('font-style', 'font-variant', 'font-weight'); + $final = ''; // output + + for ($i = 0, $size = count($bits); $i < $size; $i++) { + if ($bits[$i] === '') continue; + switch ($stage) { + + // attempting to catch font-style, font-variant or font-weight + case 0: + foreach ($stage_1 as $validator_name) { + if (isset($caught[$validator_name])) continue; + $r = $this->info[$validator_name]->validate( + $bits[$i], $config, $context); + if ($r !== false) { + $final .= $r . ' '; + $caught[$validator_name] = true; + break; + } + } + // all three caught, continue on + if (count($caught) >= 3) $stage = 1; + if ($r !== false) break; + + // attempting to catch font-size and perhaps line-height + case 1: + $found_slash = false; + if (strpos($bits[$i], '/') !== false) { + list($font_size, $line_height) = + explode('/', $bits[$i]); + if ($line_height === '') { + // ooh, there's a space after the slash! + $line_height = false; + $found_slash = true; + } + } else { + $font_size = $bits[$i]; + $line_height = false; + } + $r = $this->info['font-size']->validate( + $font_size, $config, $context); + if ($r !== false) { + $final .= $r; + // attempt to catch line-height + if ($line_height === false) { + // we need to scroll forward + for ($j = $i + 1; $j < $size; $j++) { + if ($bits[$j] === '') continue; + if ($bits[$j] === '/') { + if ($found_slash) { + return false; + } else { + $found_slash = true; + continue; + } + } + $line_height = $bits[$j]; + break; + } + } else { + // slash already found + $found_slash = true; + $j = $i; + } + if ($found_slash) { + $i = $j; + $r = $this->info['line-height']->validate( + $line_height, $config, $context); + if ($r !== false) { + $final .= '/' . $r; + } + } + $final .= ' '; + $stage = 2; + break; + } + return false; + + // attempting to catch font-family + case 2: + $font_family = + implode(' ', array_slice($bits, $i, $size - $i)); + $r = $this->info['font-family']->validate( + $font_family, $config, $context); + if ($r !== false) { + $final .= $r . ' '; + // processing completed successfully + return rtrim($final); + } + return false; + } + } + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php new file mode 100644 index 000000000..0d9a4e12c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/FontFamily.php @@ -0,0 +1,197 @@ +<?php + +/** + * Validates a font family list according to CSS spec + */ +class HTMLPurifier_AttrDef_CSS_FontFamily extends HTMLPurifier_AttrDef +{ + + protected $mask = null; + + public function __construct() { + $this->mask = '- '; + for ($c = 'a'; $c <= 'z'; $c++) $this->mask .= $c; + for ($c = 'A'; $c <= 'Z'; $c++) $this->mask .= $c; + for ($c = '0'; $c <= '9'; $c++) $this->mask .= $c; // cast-y, but should be fine + // special bytes used by UTF-8 + for ($i = 0x80; $i <= 0xFF; $i++) { + // We don't bother excluding invalid bytes in this range, + // because the our restriction of well-formed UTF-8 will + // prevent these from ever occurring. + $this->mask .= chr($i); + } + + /* + PHP's internal strcspn implementation is + O(length of string * length of mask), making it inefficient + for large masks. However, it's still faster than + preg_match 8) + for (p = s1;;) { + spanp = s2; + do { + if (*spanp == c || p == s1_end) { + return p - s1; + } + } while (spanp++ < (s2_end - 1)); + c = *++p; + } + */ + // possible optimization: invert the mask. + } + + public function validate($string, $config, $context) { + static $generic_names = array( + 'serif' => true, + 'sans-serif' => true, + 'monospace' => true, + 'fantasy' => true, + 'cursive' => true + ); + $allowed_fonts = $config->get('CSS.AllowedFonts'); + + // assume that no font names contain commas in them + $fonts = explode(',', $string); + $final = ''; + foreach($fonts as $font) { + $font = trim($font); + if ($font === '') continue; + // match a generic name + if (isset($generic_names[$font])) { + if ($allowed_fonts === null || isset($allowed_fonts[$font])) { + $final .= $font . ', '; + } + continue; + } + // match a quoted name + if ($font[0] === '"' || $font[0] === "'") { + $length = strlen($font); + if ($length <= 2) continue; + $quote = $font[0]; + if ($font[$length - 1] !== $quote) continue; + $font = substr($font, 1, $length - 2); + } + + $font = $this->expandCSSEscape($font); + + // $font is a pure representation of the font name + + if ($allowed_fonts !== null && !isset($allowed_fonts[$font])) { + continue; + } + + if (ctype_alnum($font) && $font !== '') { + // very simple font, allow it in unharmed + $final .= $font . ', '; + continue; + } + + // bugger out on whitespace. form feed (0C) really + // shouldn't show up regardless + $font = str_replace(array("\n", "\t", "\r", "\x0C"), ' ', $font); + + // Here, there are various classes of characters which need + // to be treated differently: + // - Alphanumeric characters are essentially safe. We + // handled these above. + // - Spaces require quoting, though most parsers will do + // the right thing if there aren't any characters that + // can be misinterpreted + // - Dashes rarely occur, but they fairly unproblematic + // for parsing/rendering purposes. + // The above characters cover the majority of Western font + // names. + // - Arbitrary Unicode characters not in ASCII. Because + // most parsers give little thought to Unicode, treatment + // of these codepoints is basically uniform, even for + // punctuation-like codepoints. These characters can + // show up in non-Western pages and are supported by most + // major browsers, for example: "ＭＳ 明朝" is a + // legitimate font-name + // <http://ja.wikipedia.org/wiki/MS_明朝>. See + // the CSS3 spec for more examples: + // <http://www.w3.org/TR/2011/WD-css3-fonts-20110324/localizedfamilynames.png> + // You can see live samples of these on the Internet: + // <http://www.google.co.jp/search?q=font-family+ＭＳ+明朝|ゴシック> + // However, most of these fonts have ASCII equivalents: + // for example, 'MS Mincho', and it's considered + // professional to use ASCII font names instead of + // Unicode font names. Thanks Takeshi Terada for + // providing this information. + // The following characters, to my knowledge, have not been + // used to name font names. + // - Single quote. While theoretically you might find a + // font name that has a single quote in its name (serving + // as an apostrophe, e.g. Dave's Scribble), I haven't + // been able to find any actual examples of this. + // Internet Explorer's cssText translation (which I + // believe is invoked by innerHTML) normalizes any + // quoting to single quotes, and fails to escape single + // quotes. (Note that this is not IE's behavior for all + // CSS properties, just some sort of special casing for + // font-family). So a single quote *cannot* be used + // safely in the font-family context if there will be an + // innerHTML/cssText translation. Note that Firefox 3.x + // does this too. + // - Double quote. In IE, these get normalized to + // single-quotes, no matter what the encoding. (Fun + // fact, in IE8, the 'content' CSS property gained + // support, where they special cased to preserve encoded + // double quotes, but still translate unadorned double + // quotes into single quotes.) So, because their + // fixpoint behavior is identical to single quotes, they + // cannot be allowed either. Firefox 3.x displays + // single-quote style behavior. + // - Backslashes are reduced by one (so \\ -> \) every + // iteration, so they cannot be used safely. This shows + // up in IE7, IE8 and FF3 + // - Semicolons, commas and backticks are handled properly. + // - The rest of the ASCII punctuation is handled properly. + // We haven't checked what browsers do to unadorned + // versions, but this is not important as long as the + // browser doesn't /remove/ surrounding quotes (as IE does + // for HTML). + // + // With these results in hand, we conclude that there are + // various levels of safety: + // - Paranoid: alphanumeric, spaces and dashes(?) + // - International: Paranoid + non-ASCII Unicode + // - Edgy: Everything except quotes, backslashes + // - NoJS: Standards compliance, e.g. sod IE. Note that + // with some judicious character escaping (since certain + // types of escaping doesn't work) this is theoretically + // OK as long as innerHTML/cssText is not called. + // We believe that international is a reasonable default + // (that we will implement now), and once we do more + // extensive research, we may feel comfortable with dropping + // it down to edgy. + + // Edgy: alphanumeric, spaces, dashes and Unicode. Use of + // str(c)spn assumes that the string was already well formed + // Unicode (which of course it is). + if (strspn($font, $this->mask) !== strlen($font)) { + continue; + } + + // Historical: + // In the absence of innerHTML/cssText, these ugly + // transforms don't pose a security risk (as \\ and \" + // might--these escapes are not supported by most browsers). + // We could try to be clever and use single-quote wrapping + // when there is a double quote present, but I have choosen + // not to implement that. (NOTE: you can reduce the amount + // of escapes by one depending on what quoting style you use) + // $font = str_replace('\\', '\\5C ', $font); + // $font = str_replace('"', '\\22 ', $font); + // $font = str_replace("'", '\\27 ', $font); + + // font possibly with spaces, requires quoting + $final .= "'$font', "; + } + $final = rtrim($final, ', '); + if ($final === '') return false; + return $final; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php new file mode 100644 index 000000000..779794a0b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Ident.php @@ -0,0 +1,24 @@ +<?php + +/** + * Validates based on {ident} CSS grammar production + */ +class HTMLPurifier_AttrDef_CSS_Ident extends HTMLPurifier_AttrDef +{ + + public function validate($string, $config, $context) { + + $string = trim($string); + + // early abort: '' and '0' (strings that convert to false) are invalid + if (!$string) return false; + + $pattern = '/^(-?[A-Za-z_][A-Za-z_\-0-9]*)$/'; + if (!preg_match($pattern, $string)) return false; + return $string; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php new file mode 100644 index 000000000..4e6b35e5a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ImportantDecorator.php @@ -0,0 +1,40 @@ +<?php + +/** + * Decorator which enables !important to be used in CSS values. + */ +class HTMLPurifier_AttrDef_CSS_ImportantDecorator extends HTMLPurifier_AttrDef +{ + public $def, $allow; + + /** + * @param $def Definition to wrap + * @param $allow Whether or not to allow !important + */ + public function __construct($def, $allow = false) { + $this->def = $def; + $this->allow = $allow; + } + /** + * Intercepts and removes !important if necessary + */ + public function validate($string, $config, $context) { + // test for ! and important tokens + $string = trim($string); + $is_important = false; + // :TODO: optimization: test directly for !important and ! important + if (strlen($string) >= 9 && substr($string, -9) === 'important') { + $temp = rtrim(substr($string, 0, -9)); + // use a temp, because we might want to restore important + if (strlen($temp) >= 1 && substr($temp, -1) === '!') { + $string = rtrim(substr($temp, 0, -1)); + $is_important = true; + } + } + $string = $this->def->validate($string, $config, $context); + if ($this->allow && $is_important) $string .= ' !important'; + return $string; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php new file mode 100644 index 000000000..a07ec5813 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Length.php @@ -0,0 +1,47 @@ +<?php + +/** + * Represents a Length as defined by CSS. + */ +class HTMLPurifier_AttrDef_CSS_Length extends HTMLPurifier_AttrDef +{ + + protected $min, $max; + + /** + * @param HTMLPurifier_Length $max Minimum length, or null for no bound. String is also acceptable. + * @param HTMLPurifier_Length $max Maximum length, or null for no bound. String is also acceptable. + */ + public function __construct($min = null, $max = null) { + $this->min = $min !== null ? HTMLPurifier_Length::make($min) : null; + $this->max = $max !== null ? HTMLPurifier_Length::make($max) : null; + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + + // Optimizations + if ($string === '') return false; + if ($string === '0') return '0'; + if (strlen($string) === 1) return false; + + $length = HTMLPurifier_Length::make($string); + if (!$length->isValid()) return false; + + if ($this->min) { + $c = $length->compareTo($this->min); + if ($c === false) return false; + if ($c < 0) return false; + } + if ($this->max) { + $c = $length->compareTo($this->max); + if ($c === false) return false; + if ($c > 0) return false; + } + + return $length->toString(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php new file mode 100644 index 000000000..4406868c0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/ListStyle.php @@ -0,0 +1,78 @@ +<?php + +/** + * Validates shorthand CSS property list-style. + * @warning Does not support url tokens that have internal spaces. + */ +class HTMLPurifier_AttrDef_CSS_ListStyle extends HTMLPurifier_AttrDef +{ + + /** + * Local copy of component validators. + * @note See HTMLPurifier_AttrDef_CSS_Font::$info for a similar impl. + */ + protected $info; + + public function __construct($config) { + $def = $config->getCSSDefinition(); + $this->info['list-style-type'] = $def->info['list-style-type']; + $this->info['list-style-position'] = $def->info['list-style-position']; + $this->info['list-style-image'] = $def->info['list-style-image']; + } + + public function validate($string, $config, $context) { + + // regular pre-processing + $string = $this->parseCDATA($string); + if ($string === '') return false; + + // assumes URI doesn't have spaces in it + $bits = explode(' ', strtolower($string)); // bits to process + + $caught = array(); + $caught['type'] = false; + $caught['position'] = false; + $caught['image'] = false; + + $i = 0; // number of catches + $none = false; + + foreach ($bits as $bit) { + if ($i >= 3) return; // optimization bit + if ($bit === '') continue; + foreach ($caught as $key => $status) { + if ($status !== false) continue; + $r = $this->info['list-style-' . $key]->validate($bit, $config, $context); + if ($r === false) continue; + if ($r === 'none') { + if ($none) continue; + else $none = true; + if ($key == 'image') continue; + } + $caught[$key] = $r; + $i++; + break; + } + } + + if (!$i) return false; + + $ret = array(); + + // construct type + if ($caught['type']) $ret[] = $caught['type']; + + // construct image + if ($caught['image']) $ret[] = $caught['image']; + + // construct position + if ($caught['position']) $ret[] = $caught['position']; + + if (empty($ret)) return false; + return implode(' ', $ret); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php new file mode 100644 index 000000000..4d62a40d7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Multiple.php @@ -0,0 +1,58 @@ +<?php + +/** + * Framework class for strings that involve multiple values. + * + * Certain CSS properties such as border-width and margin allow multiple + * lengths to be specified. This class can take a vanilla border-width + * definition and multiply it, usually into a max of four. + * + * @note Even though the CSS specification isn't clear about it, inherit + * can only be used alone: it will never manifest as part of a multi + * shorthand declaration. Thus, this class does not allow inherit. + */ +class HTMLPurifier_AttrDef_CSS_Multiple extends HTMLPurifier_AttrDef +{ + + /** + * Instance of component definition to defer validation to. + * @todo Make protected + */ + public $single; + + /** + * Max number of values allowed. + * @todo Make protected + */ + public $max; + + /** + * @param $single HTMLPurifier_AttrDef to multiply + * @param $max Max number of values allowed (usually four) + */ + public function __construct($single, $max = 4) { + $this->single = $single; + $this->max = $max; + } + + public function validate($string, $config, $context) { + $string = $this->parseCDATA($string); + if ($string === '') return false; + $parts = explode(' ', $string); // parseCDATA replaced \r, \t and \n + $length = count($parts); + $final = ''; + for ($i = 0, $num = 0; $i < $length && $num < $this->max; $i++) { + if (ctype_space($parts[$i])) continue; + $result = $this->single->validate($parts[$i], $config, $context); + if ($result !== false) { + $final .= $result . ' '; + $num++; + } + } + if ($final === '') return false; + return rtrim($final); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php new file mode 100644 index 000000000..3f99e12ec --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Number.php @@ -0,0 +1,69 @@ +<?php + +/** + * Validates a number as defined by the CSS spec. + */ +class HTMLPurifier_AttrDef_CSS_Number extends HTMLPurifier_AttrDef +{ + + /** + * Bool indicating whether or not only positive values allowed. + */ + protected $non_negative = false; + + /** + * @param $non_negative Bool indicating whether negatives are forbidden + */ + public function __construct($non_negative = false) { + $this->non_negative = $non_negative; + } + + /** + * @warning Some contexts do not pass $config, $context. These + * variables should not be used without checking HTMLPurifier_Length + */ + public function validate($number, $config, $context) { + + $number = $this->parseCDATA($number); + + if ($number === '') return false; + if ($number === '0') return '0'; + + $sign = ''; + switch ($number[0]) { + case '-': + if ($this->non_negative) return false; + $sign = '-'; + case '+': + $number = substr($number, 1); + } + + if (ctype_digit($number)) { + $number = ltrim($number, '0'); + return $number ? $sign . $number : '0'; + } + + // Period is the only non-numeric character allowed + if (strpos($number, '.') === false) return false; + + list($left, $right) = explode('.', $number, 2); + + if ($left === '' && $right === '') return false; + if ($left !== '' && !ctype_digit($left)) return false; + + $left = ltrim($left, '0'); + $right = rtrim($right, '0'); + + if ($right === '') { + return $left ? $sign . $left : '0'; + } elseif (!ctype_digit($right)) { + return false; + } + + return $sign . $left . '.' . $right; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php new file mode 100644 index 000000000..c34b8fc3c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/Percentage.php @@ -0,0 +1,40 @@ +<?php + +/** + * Validates a Percentage as defined by the CSS spec. + */ +class HTMLPurifier_AttrDef_CSS_Percentage extends HTMLPurifier_AttrDef +{ + + /** + * Instance of HTMLPurifier_AttrDef_CSS_Number to defer number validation + */ + protected $number_def; + + /** + * @param Bool indicating whether to forbid negative values + */ + public function __construct($non_negative = false) { + $this->number_def = new HTMLPurifier_AttrDef_CSS_Number($non_negative); + } + + public function validate($string, $config, $context) { + + $string = $this->parseCDATA($string); + + if ($string === '') return false; + $length = strlen($string); + if ($length === 1) return false; + if ($string[$length - 1] !== '%') return false; + + $number = substr($string, 0, $length - 1); + $number = $this->number_def->validate($number, $config, $context); + + if ($number === false) return false; + return "$number%"; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php new file mode 100644 index 000000000..772c922d8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/TextDecoration.php @@ -0,0 +1,38 @@ +<?php + +/** + * Validates the value for the CSS property text-decoration + * @note This class could be generalized into a version that acts sort of + * like Enum except you can compound the allowed values. + */ +class HTMLPurifier_AttrDef_CSS_TextDecoration extends HTMLPurifier_AttrDef +{ + + public function validate($string, $config, $context) { + + static $allowed_values = array( + 'line-through' => true, + 'overline' => true, + 'underline' => true, + ); + + $string = strtolower($this->parseCDATA($string)); + + if ($string === 'none') return $string; + + $parts = explode(' ', $string); + $final = ''; + foreach ($parts as $part) { + if (isset($allowed_values[$part])) { + $final .= $part . ' '; + } + } + $final = rtrim($final); + if ($final === '') return false; + return $final; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php new file mode 100644 index 000000000..c2f767e57 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/CSS/URI.php @@ -0,0 +1,61 @@ +<?php + +/** + * Validates a URI in CSS syntax, which uses url('http://example.com') + * @note While theoretically speaking a URI in a CSS document could + * be non-embedded, as of CSS2 there is no such usage so we're + * generalizing it. This may need to be changed in the future. + * @warning Since HTMLPurifier_AttrDef_CSS blindly uses semicolons as + * the separator, you cannot put a literal semicolon in + * in the URI. Try percent encoding it, in that case. + */ +class HTMLPurifier_AttrDef_CSS_URI extends HTMLPurifier_AttrDef_URI +{ + + public function __construct() { + parent::__construct(true); // always embedded + } + + public function validate($uri_string, $config, $context) { + // parse the URI out of the string and then pass it onto + // the parent object + + $uri_string = $this->parseCDATA($uri_string); + if (strpos($uri_string, 'url(') !== 0) return false; + $uri_string = substr($uri_string, 4); + $new_length = strlen($uri_string) - 1; + if ($uri_string[$new_length] != ')') return false; + $uri = trim(substr($uri_string, 0, $new_length)); + + if (!empty($uri) && ($uri[0] == "'" || $uri[0] == '"')) { + $quote = $uri[0]; + $new_length = strlen($uri) - 1; + if ($uri[$new_length] !== $quote) return false; + $uri = substr($uri, 1, $new_length - 1); + } + + $uri = $this->expandCSSEscape($uri); + + $result = parent::validate($uri, $config, $context); + + if ($result === false) return false; + + // extra sanity check; should have been done by URI + $result = str_replace(array('"', "\\", "\n", "\x0c", "\r"), "", $result); + + // suspicious characters are ()'; we're going to percent encode + // them for safety. + $result = str_replace(array('(', ')', "'"), array('%28', '%29', '%27'), $result); + + // there's an extra bug where ampersands lose their escaping on + // an innerHTML cycle, so a very unlucky query parameter could + // then change the meaning of the URL. Unfortunately, there's + // not much we can do about that... + + return "url(\"$result\")"; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Clone.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Clone.php new file mode 100644 index 000000000..ce68dbd54 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Clone.php @@ -0,0 +1,28 @@ +<?php + +/** + * Dummy AttrDef that mimics another AttrDef, BUT it generates clones + * with make. + */ +class HTMLPurifier_AttrDef_Clone extends HTMLPurifier_AttrDef +{ + /** + * What we're cloning + */ + protected $clone; + + public function __construct($clone) { + $this->clone = $clone; + } + + public function validate($v, $config, $context) { + return $this->clone->validate($v, $config, $context); + } + + public function make($string) { + return clone $this->clone; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Enum.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Enum.php new file mode 100644 index 000000000..5d603ebcc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Enum.php @@ -0,0 +1,65 @@ +<?php + +// Enum = Enumerated +/** + * Validates a keyword against a list of valid values. + * @warning The case-insensitive compare of this function uses PHP's + * built-in strtolower and ctype_lower functions, which may + * cause problems with international comparisons + */ +class HTMLPurifier_AttrDef_Enum extends HTMLPurifier_AttrDef +{ + + /** + * Lookup table of valid values. + * @todo Make protected + */ + public $valid_values = array(); + + /** + * Bool indicating whether or not enumeration is case sensitive. + * @note In general this is always case insensitive. + */ + protected $case_sensitive = false; // values according to W3C spec + + /** + * @param $valid_values List of valid values + * @param $case_sensitive Bool indicating whether or not case sensitive + */ + public function __construct( + $valid_values = array(), $case_sensitive = false + ) { + $this->valid_values = array_flip($valid_values); + $this->case_sensitive = $case_sensitive; + } + + public function validate($string, $config, $context) { + $string = trim($string); + if (!$this->case_sensitive) { + // we may want to do full case-insensitive libraries + $string = ctype_lower($string) ? $string : strtolower($string); + } + $result = isset($this->valid_values[$string]); + + return $result ? $string : false; + } + + /** + * @param $string In form of comma-delimited list of case-insensitive + * valid values. Example: "foo,bar,baz". Prepend "s:" to make + * case sensitive + */ + public function make($string) { + if (strlen($string) > 2 && $string[0] == 's' && $string[1] == ':') { + $string = substr($string, 2); + $sensitive = true; + } else { + $sensitive = false; + } + $values = explode(',', $string); + return new HTMLPurifier_AttrDef_Enum($values, $sensitive); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Bool.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Bool.php new file mode 100644 index 000000000..e06987eb8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Bool.php @@ -0,0 +1,28 @@ +<?php + +/** + * Validates a boolean attribute + */ +class HTMLPurifier_AttrDef_HTML_Bool extends HTMLPurifier_AttrDef +{ + + protected $name; + public $minimized = true; + + public function __construct($name = false) {$this->name = $name;} + + public function validate($string, $config, $context) { + if (empty($string)) return false; + return $this->name; + } + + /** + * @param $string Name of attribute + */ + public function make($string) { + return new HTMLPurifier_AttrDef_HTML_Bool($string); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Class.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Class.php new file mode 100644 index 000000000..370068d97 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Class.php @@ -0,0 +1,34 @@ +<?php + +/** + * Implements special behavior for class attribute (normally NMTOKENS) + */ +class HTMLPurifier_AttrDef_HTML_Class extends HTMLPurifier_AttrDef_HTML_Nmtokens +{ + protected function split($string, $config, $context) { + // really, this twiddle should be lazy loaded + $name = $config->getDefinition('HTML')->doctype->name; + if ($name == "XHTML 1.1" || $name == "XHTML 2.0") { + return parent::split($string, $config, $context); + } else { + return preg_split('/\s+/', $string); + } + } + protected function filter($tokens, $config, $context) { + $allowed = $config->get('Attr.AllowedClasses'); + $forbidden = $config->get('Attr.ForbiddenClasses'); + $ret = array(); + foreach ($tokens as $token) { + if ( + ($allowed === null || isset($allowed[$token])) && + !isset($forbidden[$token]) && + // We need this O(n) check because of PHP's array + // implementation that casts -0 to 0. + !in_array($token, $ret, true) + ) { + $ret[] = $token; + } + } + return $ret; + } +} diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Color.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Color.php new file mode 100644 index 000000000..00d865723 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Color.php @@ -0,0 +1,32 @@ +<?php + +/** + * Validates a color according to the HTML spec. + */ +class HTMLPurifier_AttrDef_HTML_Color extends HTMLPurifier_AttrDef +{ + + public function validate($string, $config, $context) { + + static $colors = null; + if ($colors === null) $colors = $config->get('Core.ColorKeywords'); + + $string = trim($string); + + if (empty($string)) return false; + if (isset($colors[strtolower($string)])) return $colors[$string]; + if ($string[0] === '#') $hex = substr($string, 1); + else $hex = $string; + + $length = strlen($hex); + if ($length !== 3 && $length !== 6) return false; + if (!ctype_xdigit($hex)) return false; + if ($length === 3) $hex = $hex[0].$hex[0].$hex[1].$hex[1].$hex[2].$hex[2]; + + return "#$hex"; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/FrameTarget.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/FrameTarget.php new file mode 100644 index 000000000..ae6ea7c01 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/FrameTarget.php @@ -0,0 +1,21 @@ +<?php + +/** + * Special-case enum attribute definition that lazy loads allowed frame targets + */ +class HTMLPurifier_AttrDef_HTML_FrameTarget extends HTMLPurifier_AttrDef_Enum +{ + + public $valid_values = false; // uninitialized value + protected $case_sensitive = false; + + public function __construct() {} + + public function validate($string, $config, $context) { + if ($this->valid_values === false) $this->valid_values = $config->get('Attr.AllowedFrameTargets'); + return parent::validate($string, $config, $context); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/ID.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/ID.php new file mode 100644 index 000000000..0015fa1eb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/ID.php @@ -0,0 +1,80 @@ +<?php + +/** + * Validates the HTML attribute ID. + * @warning Even though this is the id processor, it + * will ignore the directive Attr:IDBlacklist, since it will only + * go according to the ID accumulator. Since the accumulator is + * automatically generated, it will have already absorbed the + * blacklist. If you're hacking around, make sure you use load()! + */ + +class HTMLPurifier_AttrDef_HTML_ID extends HTMLPurifier_AttrDef +{ + + // selector is NOT a valid thing to use for IDREFs, because IDREFs + // *must* target IDs that exist, whereas selector #ids do not. + + /** + * Determines whether or not we're validating an ID in a CSS + * selector context. + */ + protected $selector; + + public function __construct($selector = false) { + $this->selector = $selector; + } + + public function validate($id, $config, $context) { + + if (!$this->selector && !$config->get('Attr.EnableID')) return false; + + $id = trim($id); // trim it first + + if ($id === '') return false; + + $prefix = $config->get('Attr.IDPrefix'); + if ($prefix !== '') { + $prefix .= $config->get('Attr.IDPrefixLocal'); + // prevent re-appending the prefix + if (strpos($id, $prefix) !== 0) $id = $prefix . $id; + } elseif ($config->get('Attr.IDPrefixLocal') !== '') { + trigger_error('%Attr.IDPrefixLocal cannot be used unless '. + '%Attr.IDPrefix is set', E_USER_WARNING); + } + + if (!$this->selector) { + $id_accumulator =& $context->get('IDAccumulator'); + if (isset($id_accumulator->ids[$id])) return false; + } + + // we purposely avoid using regex, hopefully this is faster + + if (ctype_alpha($id)) { + $result = true; + } else { + if (!ctype_alpha(@$id[0])) return false; + $trim = trim( // primitive style of regexps, I suppose + $id, + 'A..Za..z0..9:-._' + ); + $result = ($trim === ''); + } + + $regexp = $config->get('Attr.IDBlacklistRegexp'); + if ($regexp && preg_match($regexp, $id)) { + return false; + } + + if (!$this->selector && $result) $id_accumulator->add($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 ? $id : false; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Length.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Length.php new file mode 100644 index 000000000..a242f9c23 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Length.php @@ -0,0 +1,41 @@ +<?php + +/** + * Validates the HTML type length (not to be confused with CSS's length). + * + * This accepts integer pixels or percentages as lengths for certain + * HTML attributes. + */ + +class HTMLPurifier_AttrDef_HTML_Length extends HTMLPurifier_AttrDef_HTML_Pixels +{ + + public function validate($string, $config, $context) { + + $string = trim($string); + if ($string === '') return false; + + $parent_result = parent::validate($string, $config, $context); + if ($parent_result !== false) return $parent_result; + + $length = strlen($string); + $last_char = $string[$length - 1]; + + if ($last_char !== '%') return false; + + $points = substr($string, 0, $length - 1); + + if (!is_numeric($points)) return false; + + $points = (int) $points; + + if ($points < 0) return '0%'; + if ($points > 100) return '100%'; + + return ((string) $points) . '%'; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/LinkTypes.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/LinkTypes.php new file mode 100644 index 000000000..76d25ed08 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/LinkTypes.php @@ -0,0 +1,53 @@ +<?php + +/** + * Validates a rel/rev link attribute against a directive of allowed values + * @note We cannot use Enum because link types allow multiple + * values. + * @note Assumes link types are ASCII text + */ +class HTMLPurifier_AttrDef_HTML_LinkTypes extends HTMLPurifier_AttrDef +{ + + /** Name config attribute to pull. */ + protected $name; + + public function __construct($name) { + $configLookup = array( + 'rel' => 'AllowedRel', + 'rev' => 'AllowedRev' + ); + if (!isset($configLookup[$name])) { + trigger_error('Unrecognized attribute name for link '. + 'relationship.', E_USER_ERROR); + return; + } + $this->name = $configLookup[$name]; + } + + public function validate($string, $config, $context) { + + $allowed = $config->get('Attr.' . $this->name); + if (empty($allowed)) return false; + + $string = $this->parseCDATA($string); + $parts = explode(' ', $string); + + // lookup to prevent duplicates + $ret_lookup = array(); + foreach ($parts as $part) { + $part = strtolower(trim($part)); + if (!isset($allowed[$part])) continue; + $ret_lookup[$part] = true; + } + + if (empty($ret_lookup)) return false; + $string = implode(' ', array_keys($ret_lookup)); + + return $string; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/MultiLength.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/MultiLength.php new file mode 100644 index 000000000..c72fc76e4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/MultiLength.php @@ -0,0 +1,41 @@ +<?php + +/** + * Validates a MultiLength as defined by the HTML spec. + * + * A multilength is either a integer (pixel count), a percentage, or + * a relative number. + */ +class HTMLPurifier_AttrDef_HTML_MultiLength extends HTMLPurifier_AttrDef_HTML_Length +{ + + public function validate($string, $config, $context) { + + $string = trim($string); + if ($string === '') return false; + + $parent_result = parent::validate($string, $config, $context); + if ($parent_result !== false) return $parent_result; + + $length = strlen($string); + $last_char = $string[$length - 1]; + + if ($last_char !== '*') return false; + + $int = substr($string, 0, $length - 1); + + if ($int == '') return '*'; + if (!is_numeric($int)) return false; + + $int = (int) $int; + + if ($int < 0) return false; + if ($int == 0) return '0'; + if ($int == 1) return '*'; + return ((string) $int) . '*'; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Nmtokens.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Nmtokens.php new file mode 100644 index 000000000..aa34120bd --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Nmtokens.php @@ -0,0 +1,52 @@ +<?php + +/** + * Validates contents based on NMTOKENS attribute type. + */ +class HTMLPurifier_AttrDef_HTML_Nmtokens extends HTMLPurifier_AttrDef +{ + + public function validate($string, $config, $context) { + + $string = trim($string); + + // early abort: '' and '0' (strings that convert to false) are invalid + if (!$string) return false; + + $tokens = $this->split($string, $config, $context); + $tokens = $this->filter($tokens, $config, $context); + if (empty($tokens)) return false; + return implode(' ', $tokens); + + } + + /** + * Splits a space separated list of tokens into its constituent parts. + */ + protected function split($string, $config, $context) { + // OPTIMIZABLE! + // do the preg_match, capture all subpatterns for reformulation + + // we don't support U+00A1 and up codepoints or + // escaping because I don't know how to do that with regexps + // and plus it would complicate optimization efforts (you never + // see that anyway). + $pattern = '/(?:(?<=\s)|\A)'. // look behind for space or string start + '((?:--|-?[A-Za-z_])[A-Za-z_\-0-9]*)'. + '(?:(?=\s)|\z)/'; // look ahead for space or string end + preg_match_all($pattern, $string, $matches); + return $matches[1]; + } + + /** + * Template method for removing certain tokens based on arbitrary criteria. + * @note If we wanted to be really functional, we'd do an array_filter + * with a callback. But... we're not. + */ + protected function filter($tokens, $config, $context) { + return $tokens; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Pixels.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Pixels.php new file mode 100644 index 000000000..4cb2c1b85 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/HTML/Pixels.php @@ -0,0 +1,48 @@ +<?php + +/** + * Validates an integer representation of pixels according to the HTML spec. + */ +class HTMLPurifier_AttrDef_HTML_Pixels extends HTMLPurifier_AttrDef +{ + + protected $max; + + public function __construct($max = null) { + $this->max = $max; + } + + public function validate($string, $config, $context) { + + $string = trim($string); + if ($string === '0') return $string; + if ($string === '') return false; + $length = strlen($string); + if (substr($string, $length - 2) == 'px') { + $string = substr($string, 0, $length - 2); + } + if (!is_numeric($string)) return false; + $int = (int) $string; + + if ($int < 0) return '0'; + + // upper-bound value, extremely high values can + // crash operating systems, see <http://ha.ckers.org/imagecrash.html> + // WARNING, above link WILL crash you if you're using Windows + + if ($this->max !== null && $int > $this->max) return (string) $this->max; + + return (string) $int; + + } + + public function make($string) { + if ($string === '') $max = null; + else $max = (int) $string; + $class = get_class($this); + return new $class($max); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Integer.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Integer.php new file mode 100644 index 000000000..d59738d2a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Integer.php @@ -0,0 +1,73 @@ +<?php + +/** + * Validates an integer. + * @note While this class was modeled off the CSS definition, no currently + * allowed CSS uses this type. The properties that do are: widows, + * orphans, z-index, counter-increment, counter-reset. Some of the + * HTML attributes, however, find use for a non-negative version of this. + */ +class HTMLPurifier_AttrDef_Integer extends HTMLPurifier_AttrDef +{ + + /** + * Bool indicating whether or not negative values are allowed + */ + protected $negative = true; + + /** + * Bool indicating whether or not zero is allowed + */ + protected $zero = true; + + /** + * Bool indicating whether or not positive values are allowed + */ + protected $positive = true; + + /** + * @param $negative Bool indicating whether or not negative values are allowed + * @param $zero Bool indicating whether or not zero is allowed + * @param $positive Bool indicating whether or not positive values are allowed + */ + public function __construct( + $negative = true, $zero = true, $positive = true + ) { + $this->negative = $negative; + $this->zero = $zero; + $this->positive = $positive; + } + + public function validate($integer, $config, $context) { + + $integer = $this->parseCDATA($integer); + if ($integer === '') return false; + + // we could possibly simply typecast it to integer, but there are + // certain fringe cases that must not return an integer. + + // clip leading sign + if ( $this->negative && $integer[0] === '-' ) { + $digits = substr($integer, 1); + if ($digits === '0') $integer = '0'; // rm minus sign for zero + } elseif( $this->positive && $integer[0] === '+' ) { + $digits = $integer = substr($integer, 1); // rm unnecessary plus + } else { + $digits = $integer; + } + + // test if it's numeric + if (!ctype_digit($digits)) return false; + + // perform scope tests + if (!$this->zero && $integer == 0) return false; + if (!$this->positive && $integer > 0) return false; + if (!$this->negative && $integer < 0) return false; + + return $integer; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Lang.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Lang.php new file mode 100644 index 000000000..10e6da56d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Lang.php @@ -0,0 +1,73 @@ +<?php + +/** + * Validates the HTML attribute lang, effectively a language code. + * @note Built according to RFC 3066, which obsoleted RFC 1766 + */ +class HTMLPurifier_AttrDef_Lang extends HTMLPurifier_AttrDef +{ + + public function validate($string, $config, $context) { + + $string = trim($string); + if (!$string) return false; + + $subtags = explode('-', $string); + $num_subtags = count($subtags); + + if ($num_subtags == 0) return false; // sanity check + + // process primary subtag : $subtags[0] + $length = strlen($subtags[0]); + switch ($length) { + case 0: + return false; + case 1: + if (! ($subtags[0] == 'x' || $subtags[0] == 'i') ) { + return false; + } + break; + case 2: + case 3: + if (! ctype_alpha($subtags[0]) ) { + return false; + } elseif (! ctype_lower($subtags[0]) ) { + $subtags[0] = strtolower($subtags[0]); + } + break; + default: + return false; + } + + $new_string = $subtags[0]; + if ($num_subtags == 1) return $new_string; + + // process second subtag : $subtags[1] + $length = strlen($subtags[1]); + if ($length == 0 || ($length == 1 && $subtags[1] != 'x') || $length > 8 || !ctype_alnum($subtags[1])) { + return $new_string; + } + if (!ctype_lower($subtags[1])) $subtags[1] = strtolower($subtags[1]); + + $new_string .= '-' . $subtags[1]; + if ($num_subtags == 2) return $new_string; + + // process all other subtags, index 2 and up + for ($i = 2; $i < $num_subtags; $i++) { + $length = strlen($subtags[$i]); + if ($length == 0 || $length > 8 || !ctype_alnum($subtags[$i])) { + return $new_string; + } + if (!ctype_lower($subtags[$i])) { + $subtags[$i] = strtolower($subtags[$i]); + } + $new_string .= '-' . $subtags[$i]; + } + + return $new_string; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Switch.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Switch.php new file mode 100644 index 000000000..c9e3ed193 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Switch.php @@ -0,0 +1,34 @@ +<?php + +/** + * Decorator that, depending on a token, switches between two definitions. + */ +class HTMLPurifier_AttrDef_Switch +{ + + protected $tag; + protected $withTag, $withoutTag; + + /** + * @param string $tag Tag name to switch upon + * @param HTMLPurifier_AttrDef $with_tag Call if token matches tag + * @param HTMLPurifier_AttrDef $without_tag Call if token doesn't match, or there is no token + */ + public function __construct($tag, $with_tag, $without_tag) { + $this->tag = $tag; + $this->withTag = $with_tag; + $this->withoutTag = $without_tag; + } + + public function validate($string, $config, $context) { + $token = $context->get('CurrentToken', true); + if (!$token || $token->name !== $this->tag) { + return $this->withoutTag->validate($string, $config, $context); + } else { + return $this->withTag->validate($string, $config, $context); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Text.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Text.php new file mode 100644 index 000000000..c6216cc53 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/Text.php @@ -0,0 +1,15 @@ +<?php + +/** + * Validates arbitrary text according to the HTML spec. + */ +class HTMLPurifier_AttrDef_Text extends HTMLPurifier_AttrDef +{ + + public function validate($string, $config, $context) { + return $this->parseCDATA($string); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI.php new file mode 100644 index 000000000..c2b684671 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI.php @@ -0,0 +1,77 @@ +<?php + +/** + * Validates a URI as defined by RFC 3986. + * @note Scheme-specific mechanics deferred to HTMLPurifier_URIScheme + */ +class HTMLPurifier_AttrDef_URI extends HTMLPurifier_AttrDef +{ + + protected $parser; + protected $embedsResource; + + /** + * @param $embeds_resource_resource Does the URI here result in an extra HTTP request? + */ + public function __construct($embeds_resource = false) { + $this->parser = new HTMLPurifier_URIParser(); + $this->embedsResource = (bool) $embeds_resource; + } + + public function make($string) { + $embeds = ($string === 'embedded'); + return new HTMLPurifier_AttrDef_URI($embeds); + } + + public function validate($uri, $config, $context) { + + if ($config->get('URI.Disable')) return false; + + $uri = $this->parseCDATA($uri); + + // parse the URI + $uri = $this->parser->parse($uri); + if ($uri === false) return false; + + // add embedded flag to context for validators + $context->register('EmbeddedURI', $this->embedsResource); + + $ok = false; + do { + + // generic validation + $result = $uri->validate($config, $context); + if (!$result) break; + + // chained filtering + $uri_def = $config->getDefinition('URI'); + $result = $uri_def->filter($uri, $config, $context); + if (!$result) break; + + // scheme-specific validation + $scheme_obj = $uri->getSchemeObj($config, $context); + if (!$scheme_obj) break; + if ($this->embedsResource && !$scheme_obj->browsable) break; + $result = $scheme_obj->validate($uri, $config, $context); + if (!$result) break; + + // Post chained filtering + $result = $uri_def->postFilter($uri, $config, $context); + if (!$result) break; + + // survived gauntlet + $ok = true; + + } while (false); + + $context->destroy('EmbeddedURI'); + if (!$ok) return false; + + // back to string + return $uri->toString(); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email.php new file mode 100644 index 000000000..bfee9d166 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email.php @@ -0,0 +1,17 @@ +<?php + +abstract class HTMLPurifier_AttrDef_URI_Email extends HTMLPurifier_AttrDef +{ + + /** + * Unpacks a mailbox into its display-name and address + */ + function unpack($string) { + // needs to be implemented + } + +} + +// sub-implementations + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email/SimpleCheck.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email/SimpleCheck.php new file mode 100644 index 000000000..94c715ab4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Email/SimpleCheck.php @@ -0,0 +1,21 @@ +<?php + +/** + * Primitive email validation class based on the regexp found at + * http://www.regular-expressions.info/email.html + */ +class HTMLPurifier_AttrDef_URI_Email_SimpleCheck extends HTMLPurifier_AttrDef_URI_Email +{ + + public function validate($string, $config, $context) { + // no support for named mailboxes i.e. "Bob <bob@example.com>" + // that needs more percent encoding to be done + if ($string == '') return false; + $string = trim($string); + $result = preg_match('/^[A-Z0-9._%-]+@[A-Z0-9.-]+\.[A-Z]{2,4}$/i', $string); + return $result ? $string : false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Host.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Host.php new file mode 100644 index 000000000..125decb2d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/Host.php @@ -0,0 +1,101 @@ +<?php + +/** + * Validates a host according to the IPv4, IPv6 and DNS (future) specifications. + */ +class HTMLPurifier_AttrDef_URI_Host extends HTMLPurifier_AttrDef +{ + + /** + * Instance of HTMLPurifier_AttrDef_URI_IPv4 sub-validator + */ + protected $ipv4; + + /** + * Instance of HTMLPurifier_AttrDef_URI_IPv6 sub-validator + */ + protected $ipv6; + + public function __construct() { + $this->ipv4 = new HTMLPurifier_AttrDef_URI_IPv4(); + $this->ipv6 = new HTMLPurifier_AttrDef_URI_IPv6(); + } + + public function validate($string, $config, $context) { + $length = strlen($string); + // empty hostname is OK; it's usually semantically equivalent: + // the default host as defined by a URI scheme is used: + // + // If the URI scheme defines a default for host, then that + // default applies when the host subcomponent is undefined + // or when the registered name is empty (zero length). + if ($string === '') return ''; + if ($length > 1 && $string[0] === '[' && $string[$length-1] === ']') { + //IPv6 + $ip = substr($string, 1, $length - 2); + $valid = $this->ipv6->validate($ip, $config, $context); + if ($valid === false) return false; + return '['. $valid . ']'; + } + + // need to do checks on unusual encodings too + $ipv4 = $this->ipv4->validate($string, $config, $context); + if ($ipv4 !== false) return $ipv4; + + // A regular domain name. + + // This doesn't match I18N domain names, but we don't have proper IRI support, + // so force users to insert Punycode. + + // The productions describing this are: + $a = '[a-z]'; // alpha + $an = '[a-z0-9]'; // alphanum + $and = '[a-z0-9-]'; // alphanum | "-" + // domainlabel = alphanum | alphanum *( alphanum | "-" ) alphanum + $domainlabel = "$an($and*$an)?"; + // toplabel = alpha | alpha *( alphanum | "-" ) alphanum + $toplabel = "$a($and*$an)?"; + // hostname = *( domainlabel "." ) toplabel [ "." ] + if (preg_match("/^($domainlabel\.)*$toplabel\.?$/i", $string)) { + return $string; + } + + // If we have Net_IDNA2 support, we can support IRIs by + // punycoding them. (This is the most portable thing to do, + // since otherwise we have to assume browsers support + + if ($config->get('Core.EnableIDNA')) { + $idna = new Net_IDNA2(array('encoding' => 'utf8', 'overlong' => false, 'strict' => true)); + // we need to encode each period separately + $parts = explode('.', $string); + try { + $new_parts = array(); + foreach ($parts as $part) { + $encodable = false; + for ($i = 0, $c = strlen($part); $i < $c; $i++) { + if (ord($part[$i]) > 0x7a) { + $encodable = true; + break; + } + } + if (!$encodable) { + $new_parts[] = $part; + } else { + $new_parts[] = $idna->encode($part); + } + } + $string = implode('.', $new_parts); + if (preg_match("/^($domainlabel\.)*$toplabel\.?$/i", $string)) { + return $string; + } + } catch (Exception $e) { + // XXX error reporting + } + } + + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv4.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv4.php new file mode 100644 index 000000000..ec4cf591b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv4.php @@ -0,0 +1,39 @@ +<?php + +/** + * Validates an IPv4 address + * @author Feyd @ forums.devnetwork.net (public domain) + */ +class HTMLPurifier_AttrDef_URI_IPv4 extends HTMLPurifier_AttrDef +{ + + /** + * IPv4 regex, protected so that IPv6 can reuse it + */ + protected $ip4; + + public function validate($aIP, $config, $context) { + + if (!$this->ip4) $this->_loadRegex(); + + if (preg_match('#^' . $this->ip4 . '$#s', $aIP)) + { + return $aIP; + } + + return false; + + } + + /** + * Lazy load function to prevent regex from being stuffed in + * cache. + */ + protected function _loadRegex() { + $oct = '(?:25[0-5]|2[0-4][0-9]|1[0-9]{2}|[1-9][0-9]|[0-9])'; // 0-255 + $this->ip4 = "(?:{$oct}\\.{$oct}\\.{$oct}\\.{$oct})"; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv6.php b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv6.php new file mode 100644 index 000000000..9454e9be5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrDef/URI/IPv6.php @@ -0,0 +1,99 @@ +<?php + +/** + * Validates an IPv6 address. + * @author Feyd @ forums.devnetwork.net (public domain) + * @note This function requires brackets to have been removed from address + * in URI. + */ +class HTMLPurifier_AttrDef_URI_IPv6 extends HTMLPurifier_AttrDef_URI_IPv4 +{ + + public function validate($aIP, $config, $context) { + + if (!$this->ip4) $this->_loadRegex(); + + $original = $aIP; + + $hex = '[0-9a-fA-F]'; + $blk = '(?:' . $hex . '{1,4})'; + $pre = '(?:/(?:12[0-8]|1[0-1][0-9]|[1-9][0-9]|[0-9]))'; // /0 - /128 + + // prefix check + if (strpos($aIP, '/') !== false) + { + if (preg_match('#' . $pre . '$#s', $aIP, $find)) + { + $aIP = substr($aIP, 0, 0-strlen($find[0])); + unset($find); + } + else + { + return false; + } + } + + // IPv4-compatiblity check + if (preg_match('#(?<=:'.')' . $this->ip4 . '$#s', $aIP, $find)) + { + $aIP = substr($aIP, 0, 0-strlen($find[0])); + $ip = explode('.', $find[0]); + $ip = array_map('dechex', $ip); + $aIP .= $ip[0] . $ip[1] . ':' . $ip[2] . $ip[3]; + unset($find, $ip); + } + + // compression check + $aIP = explode('::', $aIP); + $c = count($aIP); + if ($c > 2) + { + return false; + } + elseif ($c == 2) + { + list($first, $second) = $aIP; + $first = explode(':', $first); + $second = explode(':', $second); + + if (count($first) + count($second) > 8) + { + return false; + } + + while(count($first) < 8) + { + array_push($first, '0'); + } + + array_splice($first, 8 - count($second), 8, $second); + $aIP = $first; + unset($first,$second); + } + else + { + $aIP = explode(':', $aIP[0]); + } + $c = count($aIP); + + if ($c != 8) + { + return false; + } + + // All the pieces should be 16-bit hex strings. Are they? + foreach ($aIP as $piece) + { + if (!preg_match('#^[0-9a-fA-F]{4}$#s', sprintf('%04s', $piece))) + { + return false; + } + } + + return $original; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform.php new file mode 100644 index 000000000..e61d3e01b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform.php @@ -0,0 +1,56 @@ +<?php + +/** + * Processes an entire attribute array for corrections needing multiple values. + * + * Occasionally, a certain attribute will need to be removed and popped onto + * another value. Instead of creating a complex return syntax for + * HTMLPurifier_AttrDef, we just pass the whole attribute array to a + * specialized object and have that do the special work. That is the + * family of HTMLPurifier_AttrTransform. + * + * An attribute transformation can be assigned to run before or after + * HTMLPurifier_AttrDef validation. See HTMLPurifier_HTMLDefinition for + * more details. + */ + +abstract class HTMLPurifier_AttrTransform +{ + + /** + * Abstract: makes changes to the attributes dependent on multiple values. + * + * @param $attr Assoc array of attributes, usually from + * HTMLPurifier_Token_Tag::$attr + * @param $config Mandatory HTMLPurifier_Config object. + * @param $context Mandatory HTMLPurifier_Context object + * @returns Processed attribute array. + */ + abstract public function transform($attr, $config, $context); + + /** + * Prepends CSS properties to the style attribute, creating the + * attribute if it doesn't exist. + * @param $attr Attribute array to process (passed by reference) + * @param $css CSS to prepend + */ + public function prependCSS(&$attr, $css) { + $attr['style'] = isset($attr['style']) ? $attr['style'] : ''; + $attr['style'] = $css . $attr['style']; + } + + /** + * Retrieves and removes an attribute + * @param $attr Attribute array to process (passed by reference) + * @param $key Key of attribute to confiscate + */ + public function confiscateAttr(&$attr, $key) { + if (!isset($attr[$key])) return null; + $value = $attr[$key]; + unset($attr[$key]); + return $value; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Background.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Background.php new file mode 100644 index 000000000..0e1ff24a3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Background.php @@ -0,0 +1,23 @@ +<?php + +/** + * Pre-transform that changes proprietary background attribute to CSS. + */ +class HTMLPurifier_AttrTransform_Background extends HTMLPurifier_AttrTransform { + + public function transform($attr, $config, $context) { + + if (!isset($attr['background'])) return $attr; + + $background = $this->confiscateAttr($attr, 'background'); + // some validation should happen here + + $this->prependCSS($attr, "background-image:url($background);"); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BdoDir.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BdoDir.php new file mode 100644 index 000000000..4d1a05665 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BdoDir.php @@ -0,0 +1,19 @@ +<?php + +// this MUST be placed in post, as it assumes that any value in dir is valid + +/** + * Post-trasnform that ensures that bdo tags have the dir attribute set. + */ +class HTMLPurifier_AttrTransform_BdoDir extends HTMLPurifier_AttrTransform +{ + + public function transform($attr, $config, $context) { + if (isset($attr['dir'])) return $attr; + $attr['dir'] = $config->get('Attr.DefaultTextDir'); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BgColor.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BgColor.php new file mode 100644 index 000000000..ad3916bb9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BgColor.php @@ -0,0 +1,23 @@ +<?php + +/** + * Pre-transform that changes deprecated bgcolor attribute to CSS. + */ +class HTMLPurifier_AttrTransform_BgColor extends HTMLPurifier_AttrTransform { + + public function transform($attr, $config, $context) { + + if (!isset($attr['bgcolor'])) return $attr; + + $bgcolor = $this->confiscateAttr($attr, 'bgcolor'); + // some validation should happen here + + $this->prependCSS($attr, "background-color:$bgcolor;"); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BoolToCSS.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BoolToCSS.php new file mode 100644 index 000000000..51159b671 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/BoolToCSS.php @@ -0,0 +1,36 @@ +<?php + +/** + * Pre-transform that changes converts a boolean attribute to fixed CSS + */ +class HTMLPurifier_AttrTransform_BoolToCSS extends HTMLPurifier_AttrTransform { + + /** + * Name of boolean attribute that is trigger + */ + protected $attr; + + /** + * CSS declarations to add to style, needs trailing semicolon + */ + protected $css; + + /** + * @param $attr string attribute name to convert from + * @param $css string CSS declarations to add to style (needs semicolon) + */ + public function __construct($attr, $css) { + $this->attr = $attr; + $this->css = $css; + } + + public function transform($attr, $config, $context) { + if (!isset($attr[$this->attr])) return $attr; + unset($attr[$this->attr]); + $this->prependCSS($attr, $this->css); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Border.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Border.php new file mode 100644 index 000000000..476b0b079 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Border.php @@ -0,0 +1,18 @@ +<?php + +/** + * Pre-transform that changes deprecated border attribute to CSS. + */ +class HTMLPurifier_AttrTransform_Border extends HTMLPurifier_AttrTransform { + + public function transform($attr, $config, $context) { + if (!isset($attr['border'])) return $attr; + $border_width = $this->confiscateAttr($attr, 'border'); + // some validation should happen here + $this->prependCSS($attr, "border:{$border_width}px solid;"); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/EnumToCSS.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/EnumToCSS.php new file mode 100644 index 000000000..2a5b4514a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/EnumToCSS.php @@ -0,0 +1,58 @@ +<?php + +/** + * Generic pre-transform that converts an attribute with a fixed number of + * values (enumerated) to CSS. + */ +class HTMLPurifier_AttrTransform_EnumToCSS extends HTMLPurifier_AttrTransform { + + /** + * Name of attribute to transform from + */ + protected $attr; + + /** + * Lookup array of attribute values to CSS + */ + protected $enumToCSS = array(); + + /** + * Case sensitivity of the matching + * @warning Currently can only be guaranteed to work with ASCII + * values. + */ + protected $caseSensitive = false; + + /** + * @param $attr String attribute name to transform from + * @param $enumToCSS Lookup array of attribute values to CSS + * @param $case_sensitive Boolean case sensitivity indicator, default false + */ + public function __construct($attr, $enum_to_css, $case_sensitive = false) { + $this->attr = $attr; + $this->enumToCSS = $enum_to_css; + $this->caseSensitive = (bool) $case_sensitive; + } + + public function transform($attr, $config, $context) { + + if (!isset($attr[$this->attr])) return $attr; + + $value = trim($attr[$this->attr]); + unset($attr[$this->attr]); + + if (!$this->caseSensitive) $value = strtolower($value); + + if (!isset($this->enumToCSS[$value])) { + return $attr; + } + + $this->prependCSS($attr, $this->enumToCSS[$value]); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgRequired.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgRequired.php new file mode 100644 index 000000000..7f0e4b7a5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgRequired.php @@ -0,0 +1,43 @@ +<?php + +// must be called POST validation + +/** + * Transform that supplies default values for the src and alt attributes + * in img tags, as well as prevents the img tag from being removed + * because of a missing alt tag. This needs to be registered as both + * a pre and post attribute transform. + */ +class HTMLPurifier_AttrTransform_ImgRequired extends HTMLPurifier_AttrTransform +{ + + public function transform($attr, $config, $context) { + + $src = true; + if (!isset($attr['src'])) { + if ($config->get('Core.RemoveInvalidImg')) return $attr; + $attr['src'] = $config->get('Attr.DefaultInvalidImage'); + $src = false; + } + + if (!isset($attr['alt'])) { + 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); + } else { + $attr['alt'] = $alt; + } + } else { + $attr['alt'] = $config->get('Attr.DefaultInvalidImageAlt'); + } + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgSpace.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgSpace.php new file mode 100644 index 000000000..fd84c10c3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ImgSpace.php @@ -0,0 +1,44 @@ +<?php + +/** + * Pre-transform that changes deprecated hspace and vspace attributes to CSS + */ +class HTMLPurifier_AttrTransform_ImgSpace extends HTMLPurifier_AttrTransform { + + protected $attr; + protected $css = array( + 'hspace' => array('left', 'right'), + 'vspace' => array('top', 'bottom') + ); + + public function __construct($attr) { + $this->attr = $attr; + if (!isset($this->css[$attr])) { + trigger_error(htmlspecialchars($attr) . ' is not valid space attribute'); + } + } + + public function transform($attr, $config, $context) { + + if (!isset($attr[$this->attr])) return $attr; + + $width = $this->confiscateAttr($attr, $this->attr); + // some validation could happen here + + if (!isset($this->css[$this->attr])) return $attr; + + $style = ''; + foreach ($this->css[$this->attr] as $suffix) { + $property = "margin-$suffix"; + $style .= "$property:{$width}px;"; + } + + $this->prependCSS($attr, $style); + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Input.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Input.php new file mode 100644 index 000000000..16829552d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Input.php @@ -0,0 +1,40 @@ +<?php + +/** + * Performs miscellaneous cross attribute validation and filtering for + * input elements. This is meant to be a post-transform. + */ +class HTMLPurifier_AttrTransform_Input extends HTMLPurifier_AttrTransform { + + protected $pixels; + + public function __construct() { + $this->pixels = new HTMLPurifier_AttrDef_HTML_Pixels(); + } + + public function transform($attr, $config, $context) { + if (!isset($attr['type'])) $t = 'text'; + else $t = strtolower($attr['type']); + if (isset($attr['checked']) && $t !== 'radio' && $t !== 'checkbox') { + unset($attr['checked']); + } + if (isset($attr['maxlength']) && $t !== 'text' && $t !== 'password') { + unset($attr['maxlength']); + } + if (isset($attr['size']) && $t !== 'text' && $t !== 'password') { + $result = $this->pixels->validate($attr['size'], $config, $context); + if ($result === false) unset($attr['size']); + else $attr['size'] = $result; + } + if (isset($attr['src']) && $t !== 'image') { + unset($attr['src']); + } + if (!isset($attr['value']) && ($t === 'radio' || $t === 'checkbox')) { + $attr['value'] = ''; + } + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Lang.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Lang.php new file mode 100644 index 000000000..5869e7f82 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Lang.php @@ -0,0 +1,28 @@ +<?php + +/** + * Post-transform that copies lang's value to xml:lang (and vice-versa) + * @note Theoretically speaking, this could be a pre-transform, but putting + * post is more efficient. + */ +class HTMLPurifier_AttrTransform_Lang extends HTMLPurifier_AttrTransform +{ + + public function transform($attr, $config, $context) { + + $lang = isset($attr['lang']) ? $attr['lang'] : false; + $xml_lang = isset($attr['xml:lang']) ? $attr['xml:lang'] : false; + + if ($lang !== false && $xml_lang === false) { + $attr['xml:lang'] = $lang; + } elseif ($xml_lang !== false) { + $attr['lang'] = $xml_lang; + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Length.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Length.php new file mode 100644 index 000000000..ea2f30473 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Length.php @@ -0,0 +1,27 @@ +<?php + +/** + * Class for handling width/height length attribute transformations to CSS + */ +class HTMLPurifier_AttrTransform_Length extends HTMLPurifier_AttrTransform +{ + + protected $name; + protected $cssName; + + public function __construct($name, $css_name = null) { + $this->name = $name; + $this->cssName = $css_name ? $css_name : $name; + } + + public function transform($attr, $config, $context) { + if (!isset($attr[$this->name])) return $attr; + $length = $this->confiscateAttr($attr, $this->name); + if(ctype_digit($length)) $length .= 'px'; + $this->prependCSS($attr, $this->cssName . ":$length;"); + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Name.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Name.php new file mode 100644 index 000000000..15315bc73 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Name.php @@ -0,0 +1,21 @@ +<?php + +/** + * Pre-transform that changes deprecated name attribute to ID if necessary + */ +class HTMLPurifier_AttrTransform_Name extends HTMLPurifier_AttrTransform +{ + + public function transform($attr, $config, $context) { + // Abort early if we're using relaxed definition of name + if ($config->get('HTML.Attr.Name.UseCDATA')) return $attr; + if (!isset($attr['name'])) return $attr; + $id = $this->confiscateAttr($attr, 'name'); + if ( isset($attr['id'])) return $attr; + $attr['id'] = $id; + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/NameSync.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/NameSync.php new file mode 100644 index 000000000..a95638c14 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/NameSync.php @@ -0,0 +1,27 @@ +<?php + +/** + * Post-transform that performs validation to the name attribute; if + * it is present with an equivalent id attribute, it is passed through; + * otherwise validation is performed. + */ +class HTMLPurifier_AttrTransform_NameSync extends HTMLPurifier_AttrTransform +{ + + public function __construct() { + $this->idDef = new HTMLPurifier_AttrDef_HTML_ID(); + } + + public function transform($attr, $config, $context) { + if (!isset($attr['name'])) return $attr; + $name = $attr['name']; + if (isset($attr['id']) && $attr['id'] === $name) return $attr; + $result = $this->idDef->validate($name, $config, $context); + if ($result === false) unset($attr['name']); + else $attr['name'] = $result; + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Nofollow.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Nofollow.php new file mode 100644 index 000000000..f7fb1209b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Nofollow.php @@ -0,0 +1,45 @@ +<?php + +// must be called POST validation + +/** + * Adds rel="nofollow" to all outbound links. This transform is + * only attached if Attr.Nofollow is TRUE. + */ +class HTMLPurifier_AttrTransform_Nofollow extends HTMLPurifier_AttrTransform +{ + private $parser; + + public function __construct() { + $this->parser = new HTMLPurifier_URIParser(); + } + + public function transform($attr, $config, $context) { + + if (!isset($attr['href'])) { + return $attr; + } + + // XXX Kind of inefficient + $url = $this->parser->parse($attr['href']); + $scheme = $url->getSchemeObj($config, $context); + + if ($scheme->browsable && !$url->isLocal($config, $context)) { + if (isset($attr['rel'])) { + $rels = explode(' ', $attr); + if (!in_array('nofollow', $rels)) { + $rels[] = 'nofollow'; + } + $attr['rel'] = implode(' ', $rels); + } else { + $attr['rel'] = 'nofollow'; + } + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeEmbed.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeEmbed.php new file mode 100644 index 000000000..4da449981 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeEmbed.php @@ -0,0 +1,15 @@ +<?php + +class HTMLPurifier_AttrTransform_SafeEmbed extends HTMLPurifier_AttrTransform +{ + public $name = "SafeEmbed"; + + public function transform($attr, $config, $context) { + $attr['allowscriptaccess'] = 'never'; + $attr['allownetworking'] = 'internal'; + $attr['type'] = 'application/x-shockwave-flash'; + return $attr; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeObject.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeObject.php new file mode 100644 index 000000000..1ed74898b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeObject.php @@ -0,0 +1,16 @@ +<?php + +/** + * Writes default type for all objects. Currently only supports flash. + */ +class HTMLPurifier_AttrTransform_SafeObject extends HTMLPurifier_AttrTransform +{ + public $name = "SafeObject"; + + function transform($attr, $config, $context) { + if (!isset($attr['type'])) $attr['type'] = 'application/x-shockwave-flash'; + return $attr; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeParam.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeParam.php new file mode 100644 index 000000000..bd86a7455 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/SafeParam.php @@ -0,0 +1,64 @@ +<?php + +/** + * Validates name/value pairs in param tags to be used in safe objects. This + * will only allow name values it recognizes, and pre-fill certain attributes + * with required values. + * + * @note + * This class only supports Flash. In the future, Quicktime support + * may be added. + * + * @warning + * This class expects an injector to add the necessary parameters tags. + */ +class HTMLPurifier_AttrTransform_SafeParam extends HTMLPurifier_AttrTransform +{ + public $name = "SafeParam"; + private $uri; + + public function __construct() { + $this->uri = new HTMLPurifier_AttrDef_URI(true); // embedded + $this->wmode = new HTMLPurifier_AttrDef_Enum(array('window', 'opaque', 'transparent')); + } + + public function transform($attr, $config, $context) { + // If we add support for other objects, we'll need to alter the + // transforms. + switch ($attr['name']) { + // application/x-shockwave-flash + // Keep this synchronized with Injector/SafeObject.php + case 'allowScriptAccess': + $attr['value'] = 'never'; + break; + case 'allowNetworking': + $attr['value'] = 'internal'; + break; + case 'allowFullScreen': + if ($config->get('HTML.FlashAllowFullScreen')) { + $attr['value'] = ($attr['value'] == 'true') ? 'true' : 'false'; + } else { + $attr['value'] = 'false'; + } + break; + case 'wmode': + $attr['value'] = $this->wmode->validate($attr['value'], $config, $context); + break; + case 'movie': + case 'src': + $attr['name'] = "movie"; + $attr['value'] = $this->uri->validate($attr['value'], $config, $context); + break; + case 'flashvars': + // we're going to allow arbitrary inputs to the SWF, on + // the reasoning that it could only hack the SWF, not us. + break; + // add other cases to support other param name/value pairs + default: + $attr['name'] = $attr['value'] = null; + } + return $attr; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ScriptRequired.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ScriptRequired.php new file mode 100644 index 000000000..4499050a2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/ScriptRequired.php @@ -0,0 +1,16 @@ +<?php + +/** + * Implements required attribute stipulation for <script> + */ +class HTMLPurifier_AttrTransform_ScriptRequired extends HTMLPurifier_AttrTransform +{ + public function transform($attr, $config, $context) { + if (!isset($attr['type'])) { + $attr['type'] = 'text/javascript'; + } + return $attr; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/TargetBlank.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/TargetBlank.php new file mode 100644 index 000000000..a6502c749 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/TargetBlank.php @@ -0,0 +1,38 @@ +<?php + +// must be called POST validation + +/** + * Adds target="blank" to all outbound links. This transform is + * only attached if Attr.TargetBlank is TRUE. This works regardless + * of whether or not Attr.AllowedFrameTargets + */ +class HTMLPurifier_AttrTransform_TargetBlank extends HTMLPurifier_AttrTransform +{ + private $parser; + + public function __construct() { + $this->parser = new HTMLPurifier_URIParser(); + } + + public function transform($attr, $config, $context) { + + if (!isset($attr['href'])) { + return $attr; + } + + // XXX Kind of inefficient + $url = $this->parser->parse($attr['href']); + $scheme = $url->getSchemeObj($config, $context); + + if ($scheme->browsable && !$url->isBenign($config, $context)) { + $attr['target'] = 'blank'; + } + + return $attr; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Textarea.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Textarea.php new file mode 100644 index 000000000..81ac3488b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTransform/Textarea.php @@ -0,0 +1,18 @@ +<?php + +/** + * Sets height/width defaults for <textarea> + */ +class HTMLPurifier_AttrTransform_Textarea extends HTMLPurifier_AttrTransform +{ + + public function transform($attr, $config, $context) { + // Calculated from Firefox + if (!isset($attr['cols'])) $attr['cols'] = '22'; + if (!isset($attr['rows'])) $attr['rows'] = '3'; + return $attr; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrTypes.php b/lib/htmlpurifier/library/HTMLPurifier/AttrTypes.php new file mode 100644 index 000000000..6f985ff93 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrTypes.php @@ -0,0 +1,91 @@ +<?php + +/** + * Provides lookup array of attribute types to HTMLPurifier_AttrDef objects + */ +class HTMLPurifier_AttrTypes +{ + /** + * Lookup array of attribute string identifiers to concrete implementations + */ + protected $info = array(); + + /** + * Constructs the info array, supplying default implementations for attribute + * types. + */ + public function __construct() { + // XXX This is kind of poor, since we don't actually /clone/ + // instances; instead, we use the supplied make() attribute. So, + // the underlying class must know how to deal with arguments. + // With the old implementation of Enum, that ignored its + // arguments when handling a make dispatch, the IAlign + // definition wouldn't work. + + // pseudo-types, must be instantiated via shorthand + $this->info['Enum'] = new HTMLPurifier_AttrDef_Enum(); + $this->info['Bool'] = new HTMLPurifier_AttrDef_HTML_Bool(); + + $this->info['CDATA'] = new HTMLPurifier_AttrDef_Text(); + $this->info['ID'] = new HTMLPurifier_AttrDef_HTML_ID(); + $this->info['Length'] = new HTMLPurifier_AttrDef_HTML_Length(); + $this->info['MultiLength'] = new HTMLPurifier_AttrDef_HTML_MultiLength(); + $this->info['NMTOKENS'] = new HTMLPurifier_AttrDef_HTML_Nmtokens(); + $this->info['Pixels'] = new HTMLPurifier_AttrDef_HTML_Pixels(); + $this->info['Text'] = new HTMLPurifier_AttrDef_Text(); + $this->info['URI'] = new HTMLPurifier_AttrDef_URI(); + $this->info['LanguageCode'] = new HTMLPurifier_AttrDef_Lang(); + $this->info['Color'] = new HTMLPurifier_AttrDef_HTML_Color(); + $this->info['IAlign'] = self::makeEnum('top,middle,bottom,left,right'); + $this->info['LAlign'] = self::makeEnum('top,bottom,left,right'); + $this->info['FrameTarget'] = new HTMLPurifier_AttrDef_HTML_FrameTarget(); + + // unimplemented aliases + $this->info['ContentType'] = new HTMLPurifier_AttrDef_Text(); + $this->info['ContentTypes'] = new HTMLPurifier_AttrDef_Text(); + $this->info['Charsets'] = new HTMLPurifier_AttrDef_Text(); + $this->info['Character'] = new HTMLPurifier_AttrDef_Text(); + + // "proprietary" types + $this->info['Class'] = new HTMLPurifier_AttrDef_HTML_Class(); + + // number is really a positive integer (one or more digits) + // FIXME: ^^ not always, see start and value of list items + $this->info['Number'] = new HTMLPurifier_AttrDef_Integer(false, false, true); + } + + private static function makeEnum($in) { + return new HTMLPurifier_AttrDef_Clone(new HTMLPurifier_AttrDef_Enum(explode(',', $in))); + } + + /** + * Retrieves a type + * @param $type String type name + * @return Object AttrDef for type + */ + public function get($type) { + + // determine if there is any extra info tacked on + if (strpos($type, '#') !== false) list($type, $string) = explode('#', $type, 2); + else $string = ''; + + if (!isset($this->info[$type])) { + trigger_error('Cannot retrieve undefined attribute type ' . $type, E_USER_ERROR); + return; + } + + return $this->info[$type]->make($string); + + } + + /** + * Sets a new implementation for a type + * @param $type String type name + * @param $impl Object AttrDef for type + */ + public function set($type, $impl) { + $this->info[$type] = $impl; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/AttrValidator.php b/lib/htmlpurifier/library/HTMLPurifier/AttrValidator.php new file mode 100644 index 000000000..829a0f8f2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/AttrValidator.php @@ -0,0 +1,162 @@ +<?php + +/** + * Validates the attributes of a token. Doesn't manage required attributes + * very well. The only reason we factored this out was because RemoveForeignElements + * also needed it besides ValidateAttributes. + */ +class HTMLPurifier_AttrValidator +{ + + /** + * Validates the attributes of a token, returning a modified token + * that has valid tokens + * @param $token Reference to token to validate. We require a reference + * because the operation this class performs on the token are + * not atomic, so the context CurrentToken to be updated + * throughout + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + */ + public function validateToken(&$token, &$config, $context) { + + $definition = $config->getHTMLDefinition(); + $e =& $context->get('ErrorCollector', true); + + // initialize IDAccumulator if necessary + $ok =& $context->get('IDAccumulator', true); + if (!$ok) { + $id_accumulator = HTMLPurifier_IDAccumulator::build($config, $context); + $context->register('IDAccumulator', $id_accumulator); + } + + // initialize CurrentToken if necessary + $current_token =& $context->get('CurrentToken', true); + if (!$current_token) $context->register('CurrentToken', $token); + + if ( + !$token instanceof HTMLPurifier_Token_Start && + !$token instanceof HTMLPurifier_Token_Empty + ) return $token; + + // create alias to global definition array, see also $defs + // DEFINITION CALL + $d_defs = $definition->info_global_attr; + + // don't update token until the very end, to ensure an atomic update + $attr = $token->attr; + + // do global transformations (pre) + // nothing currently utilizes this + foreach ($definition->info_attr_transform_pre as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + // do local transformations only applicable to this element (pre) + // ex. <p align="right"> to <p style="text-align:right;"> + foreach ($definition->info[$token->name]->attr_transform_pre as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + // create alias to this element's attribute definition array, see + // also $d_defs (global attribute definition array) + // DEFINITION CALL + $defs = $definition->info[$token->name]->attr; + + $attr_key = false; + $context->register('CurrentAttr', $attr_key); + + // iterate through all the attribute keypairs + // Watch out for name collisions: $key has previously been used + foreach ($attr as $attr_key => $value) { + + // call the definition + if ( isset($defs[$attr_key]) ) { + // there is a local definition defined + if ($defs[$attr_key] === false) { + // We've explicitly been told not to allow this element. + // This is usually when there's a global definition + // that must be overridden. + // Theoretically speaking, we could have a + // AttrDef_DenyAll, but this is faster! + $result = false; + } else { + // validate according to the element's definition + $result = $defs[$attr_key]->validate( + $value, $config, $context + ); + } + } elseif ( isset($d_defs[$attr_key]) ) { + // there is a global definition defined, validate according + // to the global definition + $result = $d_defs[$attr_key]->validate( + $value, $config, $context + ); + } else { + // system never heard of the attribute? DELETE! + $result = false; + } + + // put the results into effect + if ($result === false || $result === null) { + // this is a generic error message that should replaced + // with more specific ones when possible + if ($e) $e->send(E_ERROR, 'AttrValidator: Attribute removed'); + + // remove the attribute + unset($attr[$attr_key]); + } elseif (is_string($result)) { + // generally, if a substitution is happening, there + // was some sort of implicit correction going on. We'll + // delegate it to the attribute classes to say exactly what. + + // simple substitution + $attr[$attr_key] = $result; + } else { + // nothing happens + } + + // we'd also want slightly more complicated substitution + // involving an array as the return value, + // although we're not sure how colliding attributes would + // resolve (certain ones would be completely overriden, + // others would prepend themselves). + } + + $context->destroy('CurrentAttr'); + + // post transforms + + // global (error reporting untested) + foreach ($definition->info_attr_transform_post as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + // local (error reporting untested) + foreach ($definition->info[$token->name]->attr_transform_post as $transform) { + $attr = $transform->transform($o = $attr, $config, $context); + if ($e) { + if ($attr != $o) $e->send(E_NOTICE, 'AttrValidator: Attributes transformed', $o, $attr); + } + } + + $token->attr = $attr; + + // destroy CurrentToken if we made it ourselves + if (!$current_token) $context->destroy('CurrentToken'); + + } + + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Bootstrap.php b/lib/htmlpurifier/library/HTMLPurifier/Bootstrap.php new file mode 100644 index 000000000..607c5b188 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Bootstrap.php @@ -0,0 +1,104 @@ +<?php + +// constants are slow, so we use as few as possible +if (!defined('HTMLPURIFIER_PREFIX')) { + define('HTMLPURIFIER_PREFIX', realpath(dirname(__FILE__) . '/..')); +} + +// accomodations for versions earlier than 5.0.2 +// borrowed from PHP_Compat, LGPL licensed, by Aidan Lister <aidan@php.net> +if (!defined('PHP_EOL')) { + switch (strtoupper(substr(PHP_OS, 0, 3))) { + case 'WIN': + define('PHP_EOL', "\r\n"); + break; + case 'DAR': + define('PHP_EOL', "\r"); + break; + default: + define('PHP_EOL', "\n"); + } +} + +/** + * Bootstrap class that contains meta-functionality for HTML Purifier such as + * the autoload function. + * + * @note + * This class may be used without any other files from HTML Purifier. + */ +class HTMLPurifier_Bootstrap +{ + + /** + * Autoload function for HTML Purifier + * @param $class Class to load + */ + public static function autoload($class) { + $file = HTMLPurifier_Bootstrap::getPath($class); + if (!$file) return false; + // Technically speaking, it should be ok and more efficient to + // just do 'require', but Antonio Parraga reports that with + // Zend extensions such as Zend debugger and APC, this invariant + // may be broken. Since we have efficient alternatives, pay + // the cost here and avoid the bug. + require_once HTMLPURIFIER_PREFIX . '/' . $file; + return true; + } + + /** + * Returns the path for a specific class. + */ + public static function getPath($class) { + if (strncmp('HTMLPurifier', $class, 12) !== 0) return false; + // Custom implementations + if (strncmp('HTMLPurifier_Language_', $class, 22) === 0) { + $code = str_replace('_', '-', substr($class, 22)); + $file = 'HTMLPurifier/Language/classes/' . $code . '.php'; + } else { + $file = str_replace('_', '/', $class) . '.php'; + } + if (!file_exists(HTMLPURIFIER_PREFIX . '/' . $file)) return false; + return $file; + } + + /** + * "Pre-registers" our autoloader on the SPL stack. + */ + public static function registerAutoload() { + $autoload = array('HTMLPurifier_Bootstrap', 'autoload'); + if ( ($funcs = spl_autoload_functions()) === false ) { + spl_autoload_register($autoload); + } elseif (function_exists('spl_autoload_unregister')) { + $buggy = version_compare(PHP_VERSION, '5.2.11', '<'); + $compat = version_compare(PHP_VERSION, '5.1.2', '<=') && + version_compare(PHP_VERSION, '5.1.0', '>='); + foreach ($funcs as $func) { + if ($buggy && is_array($func)) { + // :TRICKY: There are some compatibility issues and some + // places where we need to error out + $reflector = new ReflectionMethod($func[0], $func[1]); + if (!$reflector->isStatic()) { + throw new Exception(' + HTML Purifier autoloader registrar is not compatible + with non-static object methods due to PHP Bug #44144; + Please do not use HTMLPurifier.autoload.php (or any + file that includes this file); instead, place the code: + spl_autoload_register(array(\'HTMLPurifier_Bootstrap\', \'autoload\')) + after your own autoloaders. + '); + } + // Suprisingly, spl_autoload_register supports the + // Class::staticMethod callback format, although call_user_func doesn't + if ($compat) $func = implode('::', $func); + } + spl_autoload_unregister($func); + } + spl_autoload_register($autoload); + foreach ($funcs as $func) spl_autoload_register($func); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/CSSDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/CSSDefinition.php new file mode 100644 index 000000000..91619f5d3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/CSSDefinition.php @@ -0,0 +1,322 @@ +<?php + +/** + * Defines allowed CSS attributes and what their values are. + * @see HTMLPurifier_HTMLDefinition + */ +class HTMLPurifier_CSSDefinition extends HTMLPurifier_Definition +{ + + public $type = 'CSS'; + + /** + * Assoc array of attribute name to definition object. + */ + public $info = array(); + + /** + * Constructs the info array. The meat of this class. + */ + protected function doSetup($config) { + + $this->info['text-align'] = new HTMLPurifier_AttrDef_Enum( + array('left', 'right', 'center', 'justify'), false); + + $border_style = + $this->info['border-bottom-style'] = + $this->info['border-right-style'] = + $this->info['border-left-style'] = + $this->info['border-top-style'] = new HTMLPurifier_AttrDef_Enum( + array('none', 'hidden', 'dotted', 'dashed', 'solid', 'double', + 'groove', 'ridge', 'inset', 'outset'), false); + + $this->info['border-style'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_style); + + $this->info['clear'] = new HTMLPurifier_AttrDef_Enum( + array('none', 'left', 'right', 'both'), false); + $this->info['float'] = new HTMLPurifier_AttrDef_Enum( + array('none', 'left', 'right'), false); + $this->info['font-style'] = new HTMLPurifier_AttrDef_Enum( + array('normal', 'italic', 'oblique'), false); + $this->info['font-variant'] = new HTMLPurifier_AttrDef_Enum( + array('normal', 'small-caps'), false); + + $uri_or_none = new HTMLPurifier_AttrDef_CSS_Composite( + array( + new HTMLPurifier_AttrDef_Enum(array('none')), + new HTMLPurifier_AttrDef_CSS_URI() + ) + ); + + $this->info['list-style-position'] = new HTMLPurifier_AttrDef_Enum( + array('inside', 'outside'), false); + $this->info['list-style-type'] = new HTMLPurifier_AttrDef_Enum( + array('disc', 'circle', 'square', 'decimal', 'lower-roman', + 'upper-roman', 'lower-alpha', 'upper-alpha', 'none'), false); + $this->info['list-style-image'] = $uri_or_none; + + $this->info['list-style'] = new HTMLPurifier_AttrDef_CSS_ListStyle($config); + + $this->info['text-transform'] = new HTMLPurifier_AttrDef_Enum( + array('capitalize', 'uppercase', 'lowercase', 'none'), false); + $this->info['color'] = new HTMLPurifier_AttrDef_CSS_Color(); + + $this->info['background-image'] = $uri_or_none; + $this->info['background-repeat'] = new HTMLPurifier_AttrDef_Enum( + array('repeat', 'repeat-x', 'repeat-y', 'no-repeat') + ); + $this->info['background-attachment'] = new HTMLPurifier_AttrDef_Enum( + array('scroll', 'fixed') + ); + $this->info['background-position'] = new HTMLPurifier_AttrDef_CSS_BackgroundPosition(); + + $border_color = + $this->info['border-top-color'] = + $this->info['border-bottom-color'] = + $this->info['border-left-color'] = + $this->info['border-right-color'] = + $this->info['background-color'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('transparent')), + new HTMLPurifier_AttrDef_CSS_Color() + )); + + $this->info['background'] = new HTMLPurifier_AttrDef_CSS_Background($config); + + $this->info['border-color'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_color); + + $border_width = + $this->info['border-top-width'] = + $this->info['border-bottom-width'] = + $this->info['border-left-width'] = + $this->info['border-right-width'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('thin', 'medium', 'thick')), + new HTMLPurifier_AttrDef_CSS_Length('0') //disallow negative + )); + + $this->info['border-width'] = new HTMLPurifier_AttrDef_CSS_Multiple($border_width); + + $this->info['letter-spacing'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('normal')), + new HTMLPurifier_AttrDef_CSS_Length() + )); + + $this->info['word-spacing'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('normal')), + new HTMLPurifier_AttrDef_CSS_Length() + )); + + $this->info['font-size'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('xx-small', 'x-small', + 'small', 'medium', 'large', 'x-large', 'xx-large', + 'larger', 'smaller')), + new HTMLPurifier_AttrDef_CSS_Percentage(), + new HTMLPurifier_AttrDef_CSS_Length() + )); + + $this->info['line-height'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('normal')), + new HTMLPurifier_AttrDef_CSS_Number(true), // no negatives + new HTMLPurifier_AttrDef_CSS_Length('0'), + new HTMLPurifier_AttrDef_CSS_Percentage(true) + )); + + $margin = + $this->info['margin-top'] = + $this->info['margin-bottom'] = + $this->info['margin-left'] = + $this->info['margin-right'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage(), + new HTMLPurifier_AttrDef_Enum(array('auto')) + )); + + $this->info['margin'] = new HTMLPurifier_AttrDef_CSS_Multiple($margin); + + // non-negative + $padding = + $this->info['padding-top'] = + $this->info['padding-bottom'] = + $this->info['padding-left'] = + $this->info['padding-right'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length('0'), + new HTMLPurifier_AttrDef_CSS_Percentage(true) + )); + + $this->info['padding'] = new HTMLPurifier_AttrDef_CSS_Multiple($padding); + + $this->info['text-indent'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage() + )); + + $trusted_wh = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length('0'), + new HTMLPurifier_AttrDef_CSS_Percentage(true), + new HTMLPurifier_AttrDef_Enum(array('auto')) + )); + $max = $config->get('CSS.MaxImgLength'); + + $this->info['width'] = + $this->info['height'] = + $max === null ? + $trusted_wh : + new HTMLPurifier_AttrDef_Switch('img', + // For img tags: + new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length('0', $max), + new HTMLPurifier_AttrDef_Enum(array('auto')) + )), + // For everyone else: + $trusted_wh + ); + + $this->info['text-decoration'] = new HTMLPurifier_AttrDef_CSS_TextDecoration(); + + $this->info['font-family'] = new HTMLPurifier_AttrDef_CSS_FontFamily(); + + // this could use specialized code + $this->info['font-weight'] = new HTMLPurifier_AttrDef_Enum( + array('normal', 'bold', 'bolder', 'lighter', '100', '200', '300', + '400', '500', '600', '700', '800', '900'), false); + + // MUST be called after other font properties, as it references + // a CSSDefinition object + $this->info['font'] = new HTMLPurifier_AttrDef_CSS_Font($config); + + // same here + $this->info['border'] = + $this->info['border-bottom'] = + $this->info['border-top'] = + $this->info['border-left'] = + $this->info['border-right'] = new HTMLPurifier_AttrDef_CSS_Border($config); + + $this->info['border-collapse'] = new HTMLPurifier_AttrDef_Enum(array( + 'collapse', 'separate')); + + $this->info['caption-side'] = new HTMLPurifier_AttrDef_Enum(array( + 'top', 'bottom')); + + $this->info['table-layout'] = new HTMLPurifier_AttrDef_Enum(array( + 'auto', 'fixed')); + + $this->info['vertical-align'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Enum(array('baseline', 'sub', 'super', + 'top', 'text-top', 'middle', 'bottom', 'text-bottom')), + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage() + )); + + $this->info['border-spacing'] = new HTMLPurifier_AttrDef_CSS_Multiple(new HTMLPurifier_AttrDef_CSS_Length(), 2); + + // partial support + $this->info['white-space'] = new HTMLPurifier_AttrDef_Enum(array('nowrap')); + + if ($config->get('CSS.Proprietary')) { + $this->doSetupProprietary($config); + } + + if ($config->get('CSS.AllowTricky')) { + $this->doSetupTricky($config); + } + + if ($config->get('CSS.Trusted')) { + $this->doSetupTrusted($config); + } + + $allow_important = $config->get('CSS.AllowImportant'); + // wrap all attr-defs with decorator that handles !important + foreach ($this->info as $k => $v) { + $this->info[$k] = new HTMLPurifier_AttrDef_CSS_ImportantDecorator($v, $allow_important); + } + + $this->setupConfigStuff($config); + } + + protected function doSetupProprietary($config) { + // Internet Explorer only scrollbar colors + $this->info['scrollbar-arrow-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-base-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-darkshadow-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-face-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-highlight-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + $this->info['scrollbar-shadow-color'] = new HTMLPurifier_AttrDef_CSS_Color(); + + // technically not proprietary, but CSS3, and no one supports it + $this->info['opacity'] = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + $this->info['-moz-opacity'] = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + $this->info['-khtml-opacity'] = new HTMLPurifier_AttrDef_CSS_AlphaValue(); + + // only opacity, for now + $this->info['filter'] = new HTMLPurifier_AttrDef_CSS_Filter(); + + } + + protected function doSetupTricky($config) { + $this->info['display'] = new HTMLPurifier_AttrDef_Enum(array( + 'inline', 'block', 'list-item', 'run-in', 'compact', + 'marker', 'table', 'inline-table', 'table-row-group', + 'table-header-group', 'table-footer-group', 'table-row', + 'table-column-group', 'table-column', 'table-cell', 'table-caption', 'none' + )); + $this->info['visibility'] = new HTMLPurifier_AttrDef_Enum(array( + 'visible', 'hidden', 'collapse' + )); + $this->info['overflow'] = new HTMLPurifier_AttrDef_Enum(array('visible', 'hidden', 'auto', 'scroll')); + } + + protected function doSetupTrusted($config) { + $this->info['position'] = new HTMLPurifier_AttrDef_Enum(array( + 'static', 'relative', 'absolute', 'fixed' + )); + $this->info['top'] = + $this->info['left'] = + $this->info['right'] = + $this->info['bottom'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_CSS_Length(), + new HTMLPurifier_AttrDef_CSS_Percentage(), + new HTMLPurifier_AttrDef_Enum(array('auto')), + )); + $this->info['z-index'] = new HTMLPurifier_AttrDef_CSS_Composite(array( + new HTMLPurifier_AttrDef_Integer(), + new HTMLPurifier_AttrDef_Enum(array('auto')), + )); + } + + /** + * Performs extra config-based processing. Based off of + * HTMLPurifier_HTMLDefinition. + * @todo Refactor duplicate elements into common class (probably using + * composition, not inheritance). + */ + protected function setupConfigStuff($config) { + + // setup allowed elements + $support = "(for information on implementing this, see the ". + "support forums) "; + $allowed_properties = $config->get('CSS.AllowedProperties'); + if ($allowed_properties !== null) { + foreach ($this->info as $name => $d) { + if(!isset($allowed_properties[$name])) unset($this->info[$name]); + unset($allowed_properties[$name]); + } + // emit errors + foreach ($allowed_properties as $name => $d) { + // :TODO: Is this htmlspecialchars() call really necessary? + $name = htmlspecialchars($name); + trigger_error("Style attribute '$name' is not supported $support", E_USER_WARNING); + } + } + + $forbidden_properties = $config->get('CSS.ForbiddenProperties'); + if ($forbidden_properties !== null) { + foreach ($this->info as $name => $d) { + if (isset($forbidden_properties[$name])) { + unset($this->info[$name]); + } + } + } + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef.php new file mode 100644 index 000000000..c5d5216da --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef.php @@ -0,0 +1,48 @@ +<?php + +/** + * Defines allowed child nodes and validates tokens against it. + */ +abstract class HTMLPurifier_ChildDef +{ + /** + * Type of child definition, usually right-most part of class name lowercase. + * Used occasionally in terms of context. + */ + public $type; + + /** + * Bool that indicates whether or not an empty array of children is okay + * + * This is necessary for redundant checking when changes affecting + * a child node may cause a parent node to now be disallowed. + */ + public $allow_empty; + + /** + * Lookup array of all elements that this definition could possibly allow + */ + public $elements = array(); + + /** + * Get lookup of tag names that should not close this element automatically. + * All other elements will do so. + */ + public function getAllowedElements($config) { + return $this->elements; + } + + /** + * Validates nodes according to definition and returns modification. + * + * @param $tokens_of_children Array of HTMLPurifier_Token + * @param $config HTMLPurifier_Config object + * @param $context HTMLPurifier_Context object + * @return bool true to leave nodes as is + * @return bool false to remove parent node + * @return array of replacement child tokens + */ + abstract public function validateChildren($tokens_of_children, $config, $context); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Chameleon.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Chameleon.php new file mode 100644 index 000000000..15c364ee3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Chameleon.php @@ -0,0 +1,48 @@ +<?php + +/** + * Definition that uses different definitions depending on context. + * + * The del and ins tags are notable because they allow different types of + * elements depending on whether or not they're in a block or inline context. + * Chameleon allows this behavior to happen by using two different + * definitions depending on context. While this somewhat generalized, + * it is specifically intended for those two tags. + */ +class HTMLPurifier_ChildDef_Chameleon extends HTMLPurifier_ChildDef +{ + + /** + * Instance of the definition object to use when inline. Usually stricter. + */ + public $inline; + + /** + * Instance of the definition object to use when block. + */ + public $block; + + public $type = 'chameleon'; + + /** + * @param $inline List of elements to allow when inline. + * @param $block List of elements to allow when block. + */ + public function __construct($inline, $block) { + $this->inline = new HTMLPurifier_ChildDef_Optional($inline); + $this->block = new HTMLPurifier_ChildDef_Optional($block); + $this->elements = $this->block->elements; + } + + public function validateChildren($tokens_of_children, $config, $context) { + if ($context->get('IsInline') === false) { + return $this->block->validateChildren( + $tokens_of_children, $config, $context); + } else { + return $this->inline->validateChildren( + $tokens_of_children, $config, $context); + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Custom.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Custom.php new file mode 100644 index 000000000..b68047b4b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Custom.php @@ -0,0 +1,90 @@ +<?php + +/** + * Custom validation class, accepts DTD child definitions + * + * @warning Currently this class is an all or nothing proposition, that is, + * it will only give a bool return value. + */ +class HTMLPurifier_ChildDef_Custom extends HTMLPurifier_ChildDef +{ + public $type = 'custom'; + public $allow_empty = false; + /** + * Allowed child pattern as defined by the DTD + */ + public $dtd_regex; + /** + * PCRE regex derived from $dtd_regex + * @private + */ + private $_pcre_regex; + /** + * @param $dtd_regex Allowed child pattern from the DTD + */ + public function __construct($dtd_regex) { + $this->dtd_regex = $dtd_regex; + $this->_compileRegex(); + } + /** + * Compiles the PCRE regex from a DTD regex ($dtd_regex to $_pcre_regex) + */ + protected function _compileRegex() { + $raw = str_replace(' ', '', $this->dtd_regex); + if ($raw{0} != '(') { + $raw = "($raw)"; + } + $el = '[#a-zA-Z0-9_.-]+'; + $reg = $raw; + + // COMPLICATED! AND MIGHT BE BUGGY! I HAVE NO CLUE WHAT I'M + // DOING! Seriously: if there's problems, please report them. + + // collect all elements into the $elements array + preg_match_all("/$el/", $reg, $matches); + foreach ($matches[0] as $match) { + $this->elements[$match] = true; + } + + // setup all elements as parentheticals with leading commas + $reg = preg_replace("/$el/", '(,\\0)', $reg); + + // remove commas when they were not solicited + $reg = preg_replace("/([^,(|]\(+),/", '\\1', $reg); + + // remove all non-paranthetical commas: they are handled by first regex + $reg = preg_replace("/,\(/", '(', $reg); + + $this->_pcre_regex = $reg; + } + public function validateChildren($tokens_of_children, $config, $context) { + $list_of_children = ''; + $nesting = 0; // depth into the nest + foreach ($tokens_of_children as $token) { + if (!empty($token->is_whitespace)) continue; + + $is_child = ($nesting == 0); // direct + + if ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + if ($is_child) { + $list_of_children .= $token->name . ','; + } + } + // add leading comma to deal with stray comma declarations + $list_of_children = ',' . rtrim($list_of_children, ','); + $okay = + preg_match( + '/^,?'.$this->_pcre_regex.'$/', + $list_of_children + ); + + return (bool) $okay; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Empty.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Empty.php new file mode 100644 index 000000000..13171f665 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Empty.php @@ -0,0 +1,20 @@ +<?php + +/** + * Definition that disallows all elements. + * @warning validateChildren() in this class is actually never called, because + * empty elements are corrected in HTMLPurifier_Strategy_MakeWellFormed + * before child definitions are parsed in earnest by + * HTMLPurifier_Strategy_FixNesting. + */ +class HTMLPurifier_ChildDef_Empty extends HTMLPurifier_ChildDef +{ + public $allow_empty = true; + public $type = 'empty'; + public function __construct() {} + public function validateChildren($tokens_of_children, $config, $context) { + return array(); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/List.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/List.php new file mode 100644 index 000000000..cdaa2893a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/List.php @@ -0,0 +1,120 @@ +<?php + +/** + * Definition for list containers ul and ol. + */ +class HTMLPurifier_ChildDef_List extends HTMLPurifier_ChildDef +{ + public $type = 'list'; + // lying a little bit, so that we can handle ul and ol ourselves + // XXX: This whole business with 'wrap' is all a bit unsatisfactory + public $elements = array('li' => true, 'ul' => true, 'ol' => true); + public function validateChildren($tokens_of_children, $config, $context) { + // Flag for subclasses + $this->whitespace = false; + + // if there are no tokens, delete parent node + if (empty($tokens_of_children)) return false; + + // the new set of children + $result = array(); + + // current depth into the nest + $nesting = 0; + + // a little sanity check to make sure it's not ALL whitespace + $all_whitespace = true; + + $seen_li = false; + $need_close_li = false; + + foreach ($tokens_of_children as $token) { + if (!empty($token->is_whitespace)) { + $result[] = $token; + continue; + } + $all_whitespace = false; // phew, we're not talking about whitespace + + if ($nesting == 1 && $need_close_li) { + $result[] = new HTMLPurifier_Token_End('li'); + $nesting--; + $need_close_li = false; + } + + $is_child = ($nesting == 0); + + if ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + if ($is_child) { + if ($token->name === 'li') { + // good + $seen_li = true; + } elseif ($token->name === 'ul' || $token->name === 'ol') { + // we want to tuck this into the previous li + $need_close_li = true; + $nesting++; + if (!$seen_li) { + // create a new li element + $result[] = new HTMLPurifier_Token_Start('li'); + } else { + // backtrack until </li> found + while(true) { + $t = array_pop($result); + if ($t instanceof HTMLPurifier_Token_End) { + // XXX actually, these invariants could very plausibly be violated + // if we are doing silly things with modifying the set of allowed elements. + // FORTUNATELY, it doesn't make a difference, since the allowed + // elements are hard-coded here! + if ($t->name !== 'li') { + trigger_error("Only li present invariant violated in List ChildDef", E_USER_ERROR); + return false; + } + break; + } elseif ($t instanceof HTMLPurifier_Token_Empty) { // bleagh + if ($t->name !== 'li') { + trigger_error("Only li present invariant violated in List ChildDef", E_USER_ERROR); + return false; + } + // XXX this should have a helper for it... + $result[] = new HTMLPurifier_Token_Start('li', $t->attr, $t->line, $t->col, $t->armor); + break; + } else { + if (!$t->is_whitespace) { + trigger_error("Only whitespace present invariant violated in List ChildDef", E_USER_ERROR); + return false; + } + } + } + } + } else { + // start wrapping (this doesn't precisely mimic + // browser behavior, but what browsers do is kind of + // hard to mimic in a standards compliant way + // XXX Actually, this has no impact in practice, + // because this gets handled earlier. Arguably, + // we should rip out all of that processing + $result[] = new HTMLPurifier_Token_Start('li'); + $nesting++; + $seen_li = true; + $need_close_li = true; + } + } + $result[] = $token; + } + if ($need_close_li) { + $result[] = new HTMLPurifier_Token_End('li'); + } + if (empty($result)) return false; + if ($all_whitespace) { + return false; + } + if ($tokens_of_children == $result) return true; + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Optional.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Optional.php new file mode 100644 index 000000000..32bcb9898 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Optional.php @@ -0,0 +1,26 @@ +<?php + +/** + * Definition that allows a set of elements, and allows no children. + * @note This is a hack to reuse code from HTMLPurifier_ChildDef_Required, + * really, one shouldn't inherit from the other. Only altered behavior + * is to overload a returned false with an array. Thus, it will never + * return false. + */ +class HTMLPurifier_ChildDef_Optional extends HTMLPurifier_ChildDef_Required +{ + public $allow_empty = true; + public $type = 'optional'; + public function validateChildren($tokens_of_children, $config, $context) { + $result = parent::validateChildren($tokens_of_children, $config, $context); + // we assume that $tokens_of_children is not modified + if ($result === false) { + if (empty($tokens_of_children)) return true; + elseif ($this->whitespace) return $tokens_of_children; + else return array(); + } + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Required.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Required.php new file mode 100644 index 000000000..4889f249b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Required.php @@ -0,0 +1,117 @@ +<?php + +/** + * Definition that allows a set of elements, but disallows empty children. + */ +class HTMLPurifier_ChildDef_Required extends HTMLPurifier_ChildDef +{ + /** + * Lookup table of allowed elements. + * @public + */ + public $elements = array(); + /** + * Whether or not the last passed node was all whitespace. + */ + protected $whitespace = false; + /** + * @param $elements List of allowed element names (lowercase). + */ + public function __construct($elements) { + if (is_string($elements)) { + $elements = str_replace(' ', '', $elements); + $elements = explode('|', $elements); + } + $keys = array_keys($elements); + if ($keys == array_keys($keys)) { + $elements = array_flip($elements); + foreach ($elements as $i => $x) { + $elements[$i] = true; + if (empty($i)) unset($elements[$i]); // remove blank + } + } + $this->elements = $elements; + } + public $allow_empty = false; + public $type = 'required'; + public function validateChildren($tokens_of_children, $config, $context) { + // Flag for subclasses + $this->whitespace = false; + + // if there are no tokens, delete parent node + if (empty($tokens_of_children)) return false; + + // the new set of children + $result = array(); + + // current depth into the nest + $nesting = 0; + + // whether or not we're deleting a node + $is_deleting = false; + + // whether or not parsed character data is allowed + // this controls whether or not we silently drop a tag + // or generate escaped HTML from it + $pcdata_allowed = isset($this->elements['#PCDATA']); + + // a little sanity check to make sure it's not ALL whitespace + $all_whitespace = true; + + // some configuration + $escape_invalid_children = $config->get('Core.EscapeInvalidChildren'); + + // generator + $gen = new HTMLPurifier_Generator($config, $context); + + foreach ($tokens_of_children as $token) { + if (!empty($token->is_whitespace)) { + $result[] = $token; + continue; + } + $all_whitespace = false; // phew, we're not talking about whitespace + + $is_child = ($nesting == 0); + + if ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + if ($is_child) { + $is_deleting = false; + if (!isset($this->elements[$token->name])) { + $is_deleting = true; + if ($pcdata_allowed && $token instanceof HTMLPurifier_Token_Text) { + $result[] = $token; + } elseif ($pcdata_allowed && $escape_invalid_children) { + $result[] = new HTMLPurifier_Token_Text( + $gen->generateFromToken($token) + ); + } + continue; + } + } + if (!$is_deleting || ($pcdata_allowed && $token instanceof HTMLPurifier_Token_Text)) { + $result[] = $token; + } elseif ($pcdata_allowed && $escape_invalid_children) { + $result[] = + new HTMLPurifier_Token_Text( + $gen->generateFromToken($token) + ); + } else { + // drop silently + } + } + if (empty($result)) return false; + if ($all_whitespace) { + $this->whitespace = true; + return false; + } + if ($tokens_of_children == $result) return true; + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/StrictBlockquote.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/StrictBlockquote.php new file mode 100644 index 000000000..dfae8a6e5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/StrictBlockquote.php @@ -0,0 +1,88 @@ +<?php + +/** + * Takes the contents of blockquote when in strict and reformats for validation. + */ +class HTMLPurifier_ChildDef_StrictBlockquote extends HTMLPurifier_ChildDef_Required +{ + protected $real_elements; + protected $fake_elements; + public $allow_empty = true; + public $type = 'strictblockquote'; + protected $init = false; + + /** + * @note We don't want MakeWellFormed to auto-close inline elements since + * they might be allowed. + */ + public function getAllowedElements($config) { + $this->init($config); + return $this->fake_elements; + } + + public function validateChildren($tokens_of_children, $config, $context) { + + $this->init($config); + + // trick the parent class into thinking it allows more + $this->elements = $this->fake_elements; + $result = parent::validateChildren($tokens_of_children, $config, $context); + $this->elements = $this->real_elements; + + if ($result === false) return array(); + if ($result === true) $result = $tokens_of_children; + + $def = $config->getHTMLDefinition(); + $block_wrap_start = new HTMLPurifier_Token_Start($def->info_block_wrapper); + $block_wrap_end = new HTMLPurifier_Token_End( $def->info_block_wrapper); + $is_inline = false; + $depth = 0; + $ret = array(); + + // assuming that there are no comment tokens + foreach ($result as $i => $token) { + $token = $result[$i]; + // ifs are nested for readability + if (!$is_inline) { + if (!$depth) { + if ( + ($token instanceof HTMLPurifier_Token_Text && !$token->is_whitespace) || + (!$token instanceof HTMLPurifier_Token_Text && !isset($this->elements[$token->name])) + ) { + $is_inline = true; + $ret[] = $block_wrap_start; + } + } + } else { + if (!$depth) { + // starting tokens have been inline text / empty + if ($token instanceof HTMLPurifier_Token_Start || $token instanceof HTMLPurifier_Token_Empty) { + if (isset($this->elements[$token->name])) { + // ended + $ret[] = $block_wrap_end; + $is_inline = false; + } + } + } + } + $ret[] = $token; + if ($token instanceof HTMLPurifier_Token_Start) $depth++; + if ($token instanceof HTMLPurifier_Token_End) $depth--; + } + if ($is_inline) $ret[] = $block_wrap_end; + return $ret; + } + + private function init($config) { + if (!$this->init) { + $def = $config->getHTMLDefinition(); + // allow all inline elements + $this->real_elements = $this->elements; + $this->fake_elements = $def->info_content_sets['Flow']; + $this->fake_elements['#PCDATA'] = true; + $this->init = true; + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Table.php b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Table.php new file mode 100644 index 000000000..9a93421a1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ChildDef/Table.php @@ -0,0 +1,227 @@ +<?php + +/** + * Definition for tables. The general idea is to extract out all of the + * essential bits, and then reconstruct it later. + * + * This is a bit confusing, because the DTDs and the W3C + * validators seem to disagree on the appropriate definition. The + * DTD claims: + * + * (CAPTION?, (COL*|COLGROUP*), THEAD?, TFOOT?, TBODY+) + * + * But actually, the HTML4 spec then has this to say: + * + * The TBODY start tag is always required except when the table + * contains only one table body and no table head or foot sections. + * The TBODY end tag may always be safely omitted. + * + * So the DTD is kind of wrong. The validator is, unfortunately, kind + * of on crack. + * + * The definition changed again in XHTML1.1; and in my opinion, this + * formulation makes the most sense. + * + * caption?, ( col* | colgroup* ), (( thead?, tfoot?, tbody+ ) | ( tr+ )) + * + * Essentially, we have two modes: thead/tfoot/tbody mode, and tr mode. + * If we encounter a thead, tfoot or tbody, we are placed in the former + * mode, and we *must* wrap any stray tr segments with a tbody. But if + * we don't run into any of them, just have tr tags is OK. + */ +class HTMLPurifier_ChildDef_Table extends HTMLPurifier_ChildDef +{ + public $allow_empty = false; + public $type = 'table'; + public $elements = array('tr' => true, 'tbody' => true, 'thead' => true, + 'tfoot' => true, 'caption' => true, 'colgroup' => true, 'col' => true); + public function __construct() {} + public function validateChildren($tokens_of_children, $config, $context) { + if (empty($tokens_of_children)) return false; + + // this ensures that the loop gets run one last time before closing + // up. It's a little bit of a hack, but it works! Just make sure you + // get rid of the token later. + $tokens_of_children[] = false; + + // only one of these elements is allowed in a table + $caption = false; + $thead = false; + $tfoot = false; + + // as many of these as you want + $cols = array(); + $content = array(); + + $nesting = 0; // current depth so we can determine nodes + $is_collecting = false; // are we globbing together tokens to package + // into one of the collectors? + $collection = array(); // collected nodes + $tag_index = 0; // the first node might be whitespace, + // so this tells us where the start tag is + $tbody_mode = false; // if true, then we need to wrap any stray + // <tr>s with a <tbody>. + + foreach ($tokens_of_children as $token) { + $is_child = ($nesting == 0); + + if ($token === false) { + // terminating sequence started + } elseif ($token instanceof HTMLPurifier_Token_Start) { + $nesting++; + } elseif ($token instanceof HTMLPurifier_Token_End) { + $nesting--; + } + + // handle node collection + if ($is_collecting) { + if ($is_child) { + // okay, let's stash the tokens away + // first token tells us the type of the collection + switch ($collection[$tag_index]->name) { + case 'tbody': + $tbody_mode = true; + case 'tr': + $content[] = $collection; + break; + case 'caption': + if ($caption !== false) break; + $caption = $collection; + break; + case 'thead': + case 'tfoot': + $tbody_mode = true; + // XXX This breaks rendering properties with + // Firefox, which never floats a <thead> to + // the top. Ever. (Our scheme will float the + // first <thead> to the top.) So maybe + // <thead>s that are not first should be + // turned into <tbody>? Very tricky, indeed. + + // access the appropriate variable, $thead or $tfoot + $var = $collection[$tag_index]->name; + if ($$var === false) { + $$var = $collection; + } else { + // Oops, there's a second one! What + // should we do? Current behavior is to + // transmutate the first and last entries into + // tbody tags, and then put into content. + // Maybe a better idea is to *attach + // it* to the existing thead or tfoot? + // We don't do this, because Firefox + // doesn't float an extra tfoot to the + // bottom like it does for the first one. + $collection[$tag_index]->name = 'tbody'; + $collection[count($collection)-1]->name = 'tbody'; + $content[] = $collection; + } + break; + case 'colgroup': + $cols[] = $collection; + break; + } + $collection = array(); + $is_collecting = false; + $tag_index = 0; + } else { + // add the node to the collection + $collection[] = $token; + } + } + + // terminate + if ($token === false) break; + + if ($is_child) { + // determine what we're dealing with + if ($token->name == 'col') { + // the only empty tag in the possie, we can handle it + // immediately + $cols[] = array_merge($collection, array($token)); + $collection = array(); + $tag_index = 0; + continue; + } + switch($token->name) { + case 'caption': + case 'colgroup': + case 'thead': + case 'tfoot': + case 'tbody': + case 'tr': + $is_collecting = true; + $collection[] = $token; + continue; + default: + if (!empty($token->is_whitespace)) { + $collection[] = $token; + $tag_index++; + } + continue; + } + } + } + + if (empty($content)) return false; + + $ret = array(); + if ($caption !== false) $ret = array_merge($ret, $caption); + if ($cols !== false) foreach ($cols as $token_array) $ret = array_merge($ret, $token_array); + if ($thead !== false) $ret = array_merge($ret, $thead); + if ($tfoot !== false) $ret = array_merge($ret, $tfoot); + + if ($tbody_mode) { + // a little tricky, since the start of the collection may be + // whitespace + $inside_tbody = false; + foreach ($content as $token_array) { + // find the starting token + foreach ($token_array as $t) { + if ($t->name === 'tr' || $t->name === 'tbody') { + break; + } + } // iterator variable carries over + if ($t->name === 'tr') { + if ($inside_tbody) { + $ret = array_merge($ret, $token_array); + } else { + $ret[] = new HTMLPurifier_Token_Start('tbody'); + $ret = array_merge($ret, $token_array); + $inside_tbody = true; + } + } elseif ($t->name === 'tbody') { + if ($inside_tbody) { + $ret[] = new HTMLPurifier_Token_End('tbody'); + $inside_tbody = false; + $ret = array_merge($ret, $token_array); + } else { + $ret = array_merge($ret, $token_array); + } + } else { + trigger_error("tr/tbody in content invariant failed in Table ChildDef", E_USER_ERROR); + } + } + if ($inside_tbody) { + $ret[] = new HTMLPurifier_Token_End('tbody'); + } + } else { + foreach ($content as $token_array) { + // invariant: everything in here is <tr>s + $ret = array_merge($ret, $token_array); + } + } + + if (!empty($collection) && $is_collecting == false){ + // grab the trailing space + $ret = array_merge($ret, $collection); + } + + array_pop($tokens_of_children); // remove phantom token + + return ($ret === $tokens_of_children) ? true : $ret; + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Config.php b/lib/htmlpurifier/library/HTMLPurifier/Config.php new file mode 100644 index 000000000..554980f22 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Config.php @@ -0,0 +1,709 @@ +<?php + +/** + * Configuration object that triggers customizable behavior. + * + * @warning This class is strongly defined: that means that the class + * will fail if an undefined directive is retrieved or set. + * + * @note Many classes that could (although many times don't) use the + * configuration object make it a mandatory parameter. This is + * because a configuration object should always be forwarded, + * otherwise, you run the risk of missing a parameter and then + * being stumped when a configuration directive doesn't work. + * + * @todo Reconsider some of the public member variables + */ +class HTMLPurifier_Config +{ + + /** + * HTML Purifier's version + */ + public $version = '4.4.0'; + + /** + * Bool indicator whether or not to automatically finalize + * the object if a read operation is done + */ + public $autoFinalize = true; + + // protected member variables + + /** + * Namespace indexed array of serials for specific namespaces (see + * getSerial() for more info). + */ + protected $serials = array(); + + /** + * Serial for entire configuration object + */ + protected $serial; + + /** + * Parser for variables + */ + protected $parser = null; + + /** + * Reference HTMLPurifier_ConfigSchema for value checking + * @note This is public for introspective purposes. Please don't + * abuse! + */ + public $def; + + /** + * Indexed array of definitions + */ + protected $definitions; + + /** + * Bool indicator whether or not config is finalized + */ + protected $finalized = false; + + /** + * Property list containing configuration directives. + */ + protected $plist; + + /** + * Whether or not a set is taking place due to an + * alias lookup. + */ + private $aliasMode; + + /** + * Set to false if you do not want line and file numbers in errors + * (useful when unit testing). This will also compress some errors + * and exceptions. + */ + public $chatty = true; + + /** + * Current lock; only gets to this namespace are allowed. + */ + private $lock; + + /** + * @param $definition HTMLPurifier_ConfigSchema that defines what directives + * are allowed. + */ + public function __construct($definition, $parent = null) { + $parent = $parent ? $parent : $definition->defaultPlist; + $this->plist = new HTMLPurifier_PropertyList($parent); + $this->def = $definition; // keep a copy around for checking + $this->parser = new HTMLPurifier_VarParser_Flexible(); + } + + /** + * Convenience constructor that creates a config object based on a mixed var + * @param mixed $config Variable that defines the state of the config + * object. Can be: a HTMLPurifier_Config() object, + * an array of directives based on loadArray(), + * or a string filename of an ini file. + * @param HTMLPurifier_ConfigSchema Schema object + * @return Configured HTMLPurifier_Config object + */ + public static function create($config, $schema = null) { + if ($config instanceof HTMLPurifier_Config) { + // pass-through + return $config; + } + if (!$schema) { + $ret = HTMLPurifier_Config::createDefault(); + } else { + $ret = new HTMLPurifier_Config($schema); + } + if (is_string($config)) $ret->loadIni($config); + elseif (is_array($config)) $ret->loadArray($config); + return $ret; + } + + /** + * Creates a new config object that inherits from a previous one. + * @param HTMLPurifier_Config $config Configuration object to inherit + * from. + * @return HTMLPurifier_Config object with $config as its parent. + */ + public static function inherit(HTMLPurifier_Config $config) { + return new HTMLPurifier_Config($config->def, $config->plist); + } + + /** + * Convenience constructor that creates a default configuration object. + * @return Default HTMLPurifier_Config object. + */ + public static function createDefault() { + $definition = HTMLPurifier_ConfigSchema::instance(); + $config = new HTMLPurifier_Config($definition); + return $config; + } + + /** + * Retreives a value from the configuration. + * @param $key String key + */ + public function get($key, $a = null) { + if ($a !== null) { + $this->triggerError("Using deprecated API: use \$config->get('$key.$a') instead", E_USER_WARNING); + $key = "$key.$a"; + } + if (!$this->finalized) $this->autoFinalize(); + if (!isset($this->def->info[$key])) { + // can't add % due to SimpleTest bug + $this->triggerError('Cannot retrieve value of undefined directive ' . htmlspecialchars($key), + E_USER_WARNING); + return; + } + if (isset($this->def->info[$key]->isAlias)) { + $d = $this->def->info[$key]; + $this->triggerError('Cannot get value from aliased directive, use real name ' . $d->key, + E_USER_ERROR); + return; + } + if ($this->lock) { + list($ns) = explode('.', $key); + if ($ns !== $this->lock) { + $this->triggerError('Cannot get value of namespace ' . $ns . ' when lock for ' . $this->lock . ' is active, this probably indicates a Definition setup method is accessing directives that are not within its namespace', E_USER_ERROR); + return; + } + } + return $this->plist->get($key); + } + + /** + * Retreives an array of directives to values from a given namespace + * @param $namespace String namespace + */ + public function getBatch($namespace) { + if (!$this->finalized) $this->autoFinalize(); + $full = $this->getAll(); + if (!isset($full[$namespace])) { + $this->triggerError('Cannot retrieve undefined namespace ' . htmlspecialchars($namespace), + E_USER_WARNING); + return; + } + return $full[$namespace]; + } + + /** + * Returns a md5 signature of a segment of the configuration object + * that uniquely identifies that particular configuration + * @note Revision is handled specially and is removed from the batch + * before processing! + * @param $namespace Namespace to get serial for + */ + public function getBatchSerial($namespace) { + if (empty($this->serials[$namespace])) { + $batch = $this->getBatch($namespace); + unset($batch['DefinitionRev']); + $this->serials[$namespace] = md5(serialize($batch)); + } + return $this->serials[$namespace]; + } + + /** + * Returns a md5 signature for the entire configuration object + * that uniquely identifies that particular configuration + */ + public function getSerial() { + if (empty($this->serial)) { + $this->serial = md5(serialize($this->getAll())); + } + return $this->serial; + } + + /** + * Retrieves all directives, organized by namespace + * @warning This is a pretty inefficient function, avoid if you can + */ + public function getAll() { + if (!$this->finalized) $this->autoFinalize(); + $ret = array(); + foreach ($this->plist->squash() as $name => $value) { + list($ns, $key) = explode('.', $name, 2); + $ret[$ns][$key] = $value; + } + return $ret; + } + + /** + * Sets a value to configuration. + * @param $key String key + * @param $value Mixed value + */ + public function set($key, $value, $a = null) { + if (strpos($key, '.') === false) { + $namespace = $key; + $directive = $value; + $value = $a; + $key = "$key.$directive"; + $this->triggerError("Using deprecated API: use \$config->set('$key', ...) instead", E_USER_NOTICE); + } else { + list($namespace) = explode('.', $key); + } + if ($this->isFinalized('Cannot set directive after finalization')) return; + if (!isset($this->def->info[$key])) { + $this->triggerError('Cannot set undefined directive ' . htmlspecialchars($key) . ' to value', + E_USER_WARNING); + return; + } + $def = $this->def->info[$key]; + + if (isset($def->isAlias)) { + if ($this->aliasMode) { + $this->triggerError('Double-aliases not allowed, please fix '. + 'ConfigSchema bug with' . $key, E_USER_ERROR); + return; + } + $this->aliasMode = true; + $this->set($def->key, $value); + $this->aliasMode = false; + $this->triggerError("$key is an alias, preferred directive name is {$def->key}", E_USER_NOTICE); + return; + } + + // Raw type might be negative when using the fully optimized form + // of stdclass, which indicates allow_null == true + $rtype = is_int($def) ? $def : $def->type; + if ($rtype < 0) { + $type = -$rtype; + $allow_null = true; + } else { + $type = $rtype; + $allow_null = isset($def->allow_null); + } + + try { + $value = $this->parser->parse($value, $type, $allow_null); + } catch (HTMLPurifier_VarParserException $e) { + $this->triggerError('Value for ' . $key . ' is of invalid type, should be ' . HTMLPurifier_VarParser::getTypeName($type), E_USER_WARNING); + return; + } + if (is_string($value) && is_object($def)) { + // resolve value alias if defined + if (isset($def->aliases[$value])) { + $value = $def->aliases[$value]; + } + // check to see if the value is allowed + if (isset($def->allowed) && !isset($def->allowed[$value])) { + $this->triggerError('Value not supported, valid values are: ' . + $this->_listify($def->allowed), E_USER_WARNING); + return; + } + } + $this->plist->set($key, $value); + + // reset definitions if the directives they depend on changed + // this is a very costly process, so it's discouraged + // with finalization + if ($namespace == 'HTML' || $namespace == 'CSS' || $namespace == 'URI') { + $this->definitions[$namespace] = null; + } + + $this->serials[$namespace] = false; + } + + /** + * Convenience function for error reporting + */ + private function _listify($lookup) { + $list = array(); + foreach ($lookup as $name => $b) $list[] = $name; + return implode(', ', $list); + } + + /** + * Retrieves object reference to the HTML definition. + * @param $raw Return a copy that has not been setup yet. Must be + * called before it's been setup, otherwise won't work. + * @param $optimized If true, this method may return null, to + * indicate that a cached version of the modified + * definition object is available and no further edits + * are necessary. Consider using + * maybeGetRawHTMLDefinition, which is more explicitly + * named, instead. + */ + public function getHTMLDefinition($raw = false, $optimized = false) { + return $this->getDefinition('HTML', $raw, $optimized); + } + + /** + * Retrieves object reference to the CSS definition + * @param $raw Return a copy that has not been setup yet. Must be + * called before it's been setup, otherwise won't work. + * @param $optimized If true, this method may return null, to + * indicate that a cached version of the modified + * definition object is available and no further edits + * are necessary. Consider using + * maybeGetRawCSSDefinition, which is more explicitly + * named, instead. + */ + public function getCSSDefinition($raw = false, $optimized = false) { + return $this->getDefinition('CSS', $raw, $optimized); + } + + /** + * Retrieves object reference to the URI definition + * @param $raw Return a copy that has not been setup yet. Must be + * called before it's been setup, otherwise won't work. + * @param $optimized If true, this method may return null, to + * indicate that a cached version of the modified + * definition object is available and no further edits + * are necessary. Consider using + * maybeGetRawURIDefinition, which is more explicitly + * named, instead. + */ + public function getURIDefinition($raw = false, $optimized = false) { + return $this->getDefinition('URI', $raw, $optimized); + } + + /** + * Retrieves a definition + * @param $type Type of definition: HTML, CSS, etc + * @param $raw Whether or not definition should be returned raw + * @param $optimized Only has an effect when $raw is true. Whether + * or not to return null if the result is already present in + * the cache. This is off by default for backwards + * compatibility reasons, but you need to do things this + * way in order to ensure that caching is done properly. + * Check out enduser-customize.html for more details. + * We probably won't ever change this default, as much as the + * maybe semantics is the "right thing to do." + */ + public function getDefinition($type, $raw = false, $optimized = false) { + if ($optimized && !$raw) { + throw new HTMLPurifier_Exception("Cannot set optimized = true when raw = false"); + } + if (!$this->finalized) $this->autoFinalize(); + // temporarily suspend locks, so we can handle recursive definition calls + $lock = $this->lock; + $this->lock = null; + $factory = HTMLPurifier_DefinitionCacheFactory::instance(); + $cache = $factory->create($type, $this); + $this->lock = $lock; + if (!$raw) { + // full definition + // --------------- + // check if definition is in memory + if (!empty($this->definitions[$type])) { + $def = $this->definitions[$type]; + // check if the definition is setup + if ($def->setup) { + return $def; + } else { + $def->setup($this); + if ($def->optimized) $cache->add($def, $this); + return $def; + } + } + // check if definition is in cache + $def = $cache->get($this); + if ($def) { + // definition in cache, save to memory and return it + $this->definitions[$type] = $def; + return $def; + } + // initialize it + $def = $this->initDefinition($type); + // set it up + $this->lock = $type; + $def->setup($this); + $this->lock = null; + // save in cache + $cache->add($def, $this); + // return it + return $def; + } else { + // raw definition + // -------------- + // check preconditions + $def = null; + if ($optimized) { + if (is_null($this->get($type . '.DefinitionID'))) { + // fatally error out if definition ID not set + throw new HTMLPurifier_Exception("Cannot retrieve raw version without specifying %$type.DefinitionID"); + } + } + if (!empty($this->definitions[$type])) { + $def = $this->definitions[$type]; + if ($def->setup && !$optimized) { + $extra = $this->chatty ? " (try moving this code block earlier in your initialization)" : ""; + throw new HTMLPurifier_Exception("Cannot retrieve raw definition after it has already been setup" . $extra); + } + if ($def->optimized === null) { + $extra = $this->chatty ? " (try flushing your cache)" : ""; + throw new HTMLPurifier_Exception("Optimization status of definition is unknown" . $extra); + } + if ($def->optimized !== $optimized) { + $msg = $optimized ? "optimized" : "unoptimized"; + $extra = $this->chatty ? " (this backtrace is for the first inconsistent call, which was for a $msg raw definition)" : ""; + throw new HTMLPurifier_Exception("Inconsistent use of optimized and unoptimized raw definition retrievals" . $extra); + } + } + // check if definition was in memory + if ($def) { + if ($def->setup) { + // invariant: $optimized === true (checked above) + return null; + } else { + return $def; + } + } + // if optimized, check if definition was in cache + // (because we do the memory check first, this formulation + // is prone to cache slamming, but I think + // guaranteeing that either /all/ of the raw + // setup code or /none/ of it is run is more important.) + if ($optimized) { + // This code path only gets run once; once we put + // something in $definitions (which is guaranteed by the + // trailing code), we always short-circuit above. + $def = $cache->get($this); + if ($def) { + // save the full definition for later, but don't + // return it yet + $this->definitions[$type] = $def; + return null; + } + } + // check invariants for creation + if (!$optimized) { + if (!is_null($this->get($type . '.DefinitionID'))) { + if ($this->chatty) { + $this->triggerError("Due to a documentation error in previous version of HTML Purifier, your definitions are not being cached. If this is OK, you can remove the %$type.DefinitionRev and %$type.DefinitionID declaration. Otherwise, modify your code to use maybeGetRawDefinition, and test if the returned value is null before making any edits (if it is null, that means that a cached version is available, and no raw operations are necessary). See <a href='http://htmlpurifier.org/docs/enduser-customize.html#optimized'>Customize</a> for more details", E_USER_WARNING); + } else { + $this->triggerError("Useless DefinitionID declaration", E_USER_WARNING); + } + } + } + // initialize it + $def = $this->initDefinition($type); + $def->optimized = $optimized; + return $def; + } + throw new HTMLPurifier_Exception("The impossible happened!"); + } + + private function initDefinition($type) { + // quick checks failed, let's create the object + if ($type == 'HTML') { + $def = new HTMLPurifier_HTMLDefinition(); + } elseif ($type == 'CSS') { + $def = new HTMLPurifier_CSSDefinition(); + } elseif ($type == 'URI') { + $def = new HTMLPurifier_URIDefinition(); + } else { + throw new HTMLPurifier_Exception("Definition of $type type not supported"); + } + $this->definitions[$type] = $def; + return $def; + } + + public function maybeGetRawDefinition($name) { + return $this->getDefinition($name, true, true); + } + + public function maybeGetRawHTMLDefinition() { + return $this->getDefinition('HTML', true, true); + } + + public function maybeGetRawCSSDefinition() { + return $this->getDefinition('CSS', true, true); + } + + public function maybeGetRawURIDefinition() { + return $this->getDefinition('URI', true, true); + } + + /** + * Loads configuration values from an array with the following structure: + * Namespace.Directive => Value + * @param $config_array Configuration associative array + */ + public function loadArray($config_array) { + if ($this->isFinalized('Cannot load directives after finalization')) return; + foreach ($config_array as $key => $value) { + $key = str_replace('_', '.', $key); + if (strpos($key, '.') !== false) { + $this->set($key, $value); + } else { + $namespace = $key; + $namespace_values = $value; + foreach ($namespace_values as $directive => $value) { + $this->set($namespace .'.'. $directive, $value); + } + } + } + } + + /** + * Returns a list of array(namespace, directive) for all directives + * that are allowed in a web-form context as per an allowed + * namespaces/directives list. + * @param $allowed List of allowed namespaces/directives + */ + public static function getAllowedDirectivesForForm($allowed, $schema = null) { + if (!$schema) { + $schema = HTMLPurifier_ConfigSchema::instance(); + } + if ($allowed !== true) { + if (is_string($allowed)) $allowed = array($allowed); + $allowed_ns = array(); + $allowed_directives = array(); + $blacklisted_directives = array(); + foreach ($allowed as $ns_or_directive) { + if (strpos($ns_or_directive, '.') !== false) { + // directive + if ($ns_or_directive[0] == '-') { + $blacklisted_directives[substr($ns_or_directive, 1)] = true; + } else { + $allowed_directives[$ns_or_directive] = true; + } + } else { + // namespace + $allowed_ns[$ns_or_directive] = true; + } + } + } + $ret = array(); + foreach ($schema->info as $key => $def) { + list($ns, $directive) = explode('.', $key, 2); + if ($allowed !== true) { + if (isset($blacklisted_directives["$ns.$directive"])) continue; + if (!isset($allowed_directives["$ns.$directive"]) && !isset($allowed_ns[$ns])) continue; + } + if (isset($def->isAlias)) continue; + if ($directive == 'DefinitionID' || $directive == 'DefinitionRev') continue; + $ret[] = array($ns, $directive); + } + return $ret; + } + + /** + * Loads configuration values from $_GET/$_POST that were posted + * via ConfigForm + * @param $array $_GET or $_POST array to import + * @param $index Index/name that the config variables are in + * @param $allowed List of allowed namespaces/directives + * @param $mq_fix Boolean whether or not to enable magic quotes fix + * @param $schema Instance of HTMLPurifier_ConfigSchema to use, if not global copy + */ + public static function loadArrayFromForm($array, $index = false, $allowed = true, $mq_fix = true, $schema = null) { + $ret = HTMLPurifier_Config::prepareArrayFromForm($array, $index, $allowed, $mq_fix, $schema); + $config = HTMLPurifier_Config::create($ret, $schema); + return $config; + } + + /** + * Merges in configuration values from $_GET/$_POST to object. NOT STATIC. + * @note Same parameters as loadArrayFromForm + */ + public function mergeArrayFromForm($array, $index = false, $allowed = true, $mq_fix = true) { + $ret = HTMLPurifier_Config::prepareArrayFromForm($array, $index, $allowed, $mq_fix, $this->def); + $this->loadArray($ret); + } + + /** + * Prepares an array from a form into something usable for the more + * strict parts of HTMLPurifier_Config + */ + public static function prepareArrayFromForm($array, $index = false, $allowed = true, $mq_fix = true, $schema = null) { + if ($index !== false) $array = (isset($array[$index]) && is_array($array[$index])) ? $array[$index] : array(); + $mq = $mq_fix && function_exists('get_magic_quotes_gpc') && get_magic_quotes_gpc(); + + $allowed = HTMLPurifier_Config::getAllowedDirectivesForForm($allowed, $schema); + $ret = array(); + foreach ($allowed as $key) { + list($ns, $directive) = $key; + $skey = "$ns.$directive"; + if (!empty($array["Null_$skey"])) { + $ret[$ns][$directive] = null; + continue; + } + if (!isset($array[$skey])) continue; + $value = $mq ? stripslashes($array[$skey]) : $array[$skey]; + $ret[$ns][$directive] = $value; + } + return $ret; + } + + /** + * Loads configuration values from an ini file + * @param $filename Name of ini file + */ + public function loadIni($filename) { + if ($this->isFinalized('Cannot load directives after finalization')) return; + $array = parse_ini_file($filename, true); + $this->loadArray($array); + } + + /** + * Checks whether or not the configuration object is finalized. + * @param $error String error message, or false for no error + */ + public function isFinalized($error = false) { + if ($this->finalized && $error) { + $this->triggerError($error, E_USER_ERROR); + } + return $this->finalized; + } + + /** + * Finalizes configuration only if auto finalize is on and not + * already finalized + */ + public function autoFinalize() { + if ($this->autoFinalize) { + $this->finalize(); + } else { + $this->plist->squash(true); + } + } + + /** + * Finalizes a configuration object, prohibiting further change + */ + public function finalize() { + $this->finalized = true; + $this->parser = null; + } + + /** + * Produces a nicely formatted error message by supplying the + * stack frame information OUTSIDE of HTMLPurifier_Config. + */ + protected function triggerError($msg, $no) { + // determine previous stack frame + $extra = ''; + if ($this->chatty) { + $trace = debug_backtrace(); + // zip(tail(trace), trace) -- but PHP is not Haskell har har + for ($i = 0, $c = count($trace); $i < $c - 1; $i++) { + if ($trace[$i + 1]['class'] === 'HTMLPurifier_Config') { + continue; + } + $frame = $trace[$i]; + $extra = " invoked on line {$frame['line']} in file {$frame['file']}"; + break; + } + } + trigger_error($msg . $extra, $no); + } + + /** + * Returns a serialized form of the configuration object that can + * be reconstituted. + */ + public function serialize() { + $this->getDefinition('HTML'); + $this->getDefinition('CSS'); + $this->getDefinition('URI'); + return serialize($this); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema.php new file mode 100644 index 000000000..fadf7a589 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema.php @@ -0,0 +1,164 @@ +<?php + +/** + * Configuration definition, defines directives and their defaults. + */ +class HTMLPurifier_ConfigSchema { + + /** + * Defaults of the directives and namespaces. + * @note This shares the exact same structure as HTMLPurifier_Config::$conf + */ + public $defaults = array(); + + /** + * The default property list. Do not edit this property list. + */ + public $defaultPlist; + + /** + * Definition of the directives. The structure of this is: + * + * array( + * 'Namespace' => array( + * 'Directive' => new stdclass(), + * ) + * ) + * + * The stdclass may have the following properties: + * + * - If isAlias isn't set: + * - type: Integer type of directive, see HTMLPurifier_VarParser for definitions + * - allow_null: If set, this directive allows null values + * - aliases: If set, an associative array of value aliases to real values + * - allowed: If set, a lookup array of allowed (string) values + * - If isAlias is set: + * - 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 + * property set to the integer. If the integer is negative, type is + * equal to the absolute value of integer, and allow_null is true. + * + * This class is friendly with HTMLPurifier_Config. If you need introspection + * about the schema, you're better of using the ConfigSchema_Interchange, + * which uses more memory but has much richer information. + */ + public $info = array(); + + /** + * Application-wide singleton + */ + static protected $singleton; + + public function __construct() { + $this->defaultPlist = new HTMLPurifier_PropertyList(); + } + + /** + * Unserializes the default ConfigSchema. + */ + public static function makeFromSerial() { + $contents = file_get_contents(HTMLPURIFIER_PREFIX . '/HTMLPurifier/ConfigSchema/schema.ser'); + $r = unserialize($contents); + if (!$r) { + $hash = sha1($contents); + trigger_error("Unserialization of configuration schema failed, sha1 of file was $hash", E_USER_ERROR); + } + return $r; + } + + /** + * Retrieves an instance of the application-wide configuration definition. + */ + public static function instance($prototype = null) { + if ($prototype !== null) { + HTMLPurifier_ConfigSchema::$singleton = $prototype; + } elseif (HTMLPurifier_ConfigSchema::$singleton === null || $prototype === true) { + HTMLPurifier_ConfigSchema::$singleton = HTMLPurifier_ConfigSchema::makeFromSerial(); + } + return HTMLPurifier_ConfigSchema::$singleton; + } + + /** + * Defines a directive for configuration + * @warning Will fail of directive's namespace is defined. + * @warning This method's signature is slightly different from the legacy + * define() static method! Beware! + * @param $namespace Namespace the directive is in + * @param $name Key of directive + * @param $default Default value of directive + * @param $type Allowed type of the directive. See + * HTMLPurifier_DirectiveDef::$type for allowed values + * @param $allow_null Whether or not to allow null values + */ + public function add($key, $default, $type, $allow_null) { + $obj = new stdclass(); + $obj->type = is_int($type) ? $type : HTMLPurifier_VarParser::$types[$type]; + if ($allow_null) $obj->allow_null = true; + $this->info[$key] = $obj; + $this->defaults[$key] = $default; + $this->defaultPlist->set($key, $default); + } + + /** + * Defines a directive value alias. + * + * Directive value aliases are convenient for developers because it lets + * them set a directive to several values and get the same result. + * @param $namespace Directive's namespace + * @param $name Name of Directive + * @param $aliases Hash of aliased values to the real alias + */ + public function addValueAliases($key, $aliases) { + if (!isset($this->info[$key]->aliases)) { + $this->info[$key]->aliases = array(); + } + foreach ($aliases as $alias => $real) { + $this->info[$key]->aliases[$alias] = $real; + } + } + + /** + * Defines a set of allowed values for a directive. + * @warning This is slightly different from the corresponding static + * method definition. + * @param $namespace Namespace of directive + * @param $name Name of directive + * @param $allowed Lookup array of allowed values + */ + public function addAllowedValues($key, $allowed) { + $this->info[$key]->allowed = $allowed; + } + + /** + * Defines a directive alias for backwards compatibility + * @param $namespace + * @param $name Directive that will be aliased + * @param $new_namespace + * @param $new_name Directive that the alias will be to + */ + public function addAlias($key, $new_key) { + $obj = new stdclass; + $obj->key = $new_key; + $obj->isAlias = true; + $this->info[$key] = $obj; + } + + /** + * Replaces any stdclass that only has the type property with type integer. + */ + public function postProcess() { + foreach ($this->info as $key => $v) { + if (count((array) $v) == 1) { + $this->info[$key] = $v->type; + } elseif (count((array) $v) == 2 && isset($v->allow_null)) { + $this->info[$key] = -$v->type; + } + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Builder/ConfigSchema.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Builder/ConfigSchema.php new file mode 100644 index 000000000..c05668a70 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Builder/ConfigSchema.php @@ -0,0 +1,44 @@ +<?php + +/** + * Converts HTMLPurifier_ConfigSchema_Interchange to our runtime + * representation used to perform checks on user configuration. + */ +class HTMLPurifier_ConfigSchema_Builder_ConfigSchema +{ + + public function build($interchange) { + $schema = new HTMLPurifier_ConfigSchema(); + foreach ($interchange->directives as $d) { + $schema->add( + $d->id->key, + $d->default, + $d->type, + $d->typeAllowsNull + ); + if ($d->allowed !== null) { + $schema->addAllowedValues( + $d->id->key, + $d->allowed + ); + } + foreach ($d->aliases as $alias) { + $schema->addAlias( + $alias->key, + $d->id->key + ); + } + if ($d->valueAliases !== null) { + $schema->addValueAliases( + $d->id->key, + $d->valueAliases + ); + } + } + $schema->postProcess(); + return $schema; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Builder/Xml.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Builder/Xml.php new file mode 100644 index 000000000..244561a37 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Builder/Xml.php @@ -0,0 +1,106 @@ +<?php + +/** + * Converts HTMLPurifier_ConfigSchema_Interchange to an XML format, + * which can be further processed to generate documentation. + */ +class HTMLPurifier_ConfigSchema_Builder_Xml extends XMLWriter +{ + + protected $interchange; + private $namespace; + + protected function writeHTMLDiv($html) { + $this->startElement('div'); + + $purifier = HTMLPurifier::getInstance(); + $html = $purifier->purify($html); + $this->writeAttribute('xmlns', 'http://www.w3.org/1999/xhtml'); + $this->writeRaw($html); + + $this->endElement(); // div + } + + protected function export($var) { + if ($var === array()) return 'array()'; + return var_export($var, true); + } + + public function build($interchange) { + // global access, only use as last resort + $this->interchange = $interchange; + + $this->setIndent(true); + $this->startDocument('1.0', 'UTF-8'); + $this->startElement('configdoc'); + $this->writeElement('title', $interchange->name); + + foreach ($interchange->directives as $directive) { + $this->buildDirective($directive); + } + + if ($this->namespace) $this->endElement(); // namespace + + $this->endElement(); // configdoc + $this->flush(); + } + + public function buildDirective($directive) { + + // Kludge, although I suppose having a notion of a "root namespace" + // certainly makes things look nicer when documentation is built. + // Depends on things being sorted. + if (!$this->namespace || $this->namespace !== $directive->id->getRootNamespace()) { + if ($this->namespace) $this->endElement(); // namespace + $this->namespace = $directive->id->getRootNamespace(); + $this->startElement('namespace'); + $this->writeAttribute('id', $this->namespace); + $this->writeElement('name', $this->namespace); + } + + $this->startElement('directive'); + $this->writeAttribute('id', $directive->id->toString()); + + $this->writeElement('name', $directive->id->getDirective()); + + $this->startElement('aliases'); + foreach ($directive->aliases as $alias) $this->writeElement('alias', $alias->toString()); + $this->endElement(); // aliases + + $this->startElement('constraints'); + if ($directive->version) $this->writeElement('version', $directive->version); + $this->startElement('type'); + if ($directive->typeAllowsNull) $this->writeAttribute('allow-null', 'yes'); + $this->text($directive->type); + $this->endElement(); // type + if ($directive->allowed) { + $this->startElement('allowed'); + foreach ($directive->allowed as $value => $x) $this->writeElement('value', $value); + $this->endElement(); // allowed + } + $this->writeElement('default', $this->export($directive->default)); + $this->writeAttribute('xml:space', 'preserve'); + if ($directive->external) { + $this->startElement('external'); + foreach ($directive->external as $project) $this->writeElement('project', $project); + $this->endElement(); + } + $this->endElement(); // constraints + + if ($directive->deprecatedVersion) { + $this->startElement('deprecated'); + $this->writeElement('version', $directive->deprecatedVersion); + $this->writeElement('use', $directive->deprecatedUse->toString()); + $this->endElement(); // deprecated + } + + $this->startElement('description'); + $this->writeHTMLDiv($directive->description); + $this->endElement(); // description + + $this->endElement(); // directive + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Exception.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Exception.php new file mode 100644 index 000000000..2671516c5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Exception.php @@ -0,0 +1,11 @@ +<?php + +/** + * Exceptions related to configuration schema + */ +class HTMLPurifier_ConfigSchema_Exception extends HTMLPurifier_Exception +{ + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange.php new file mode 100644 index 000000000..91a5aa730 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange.php @@ -0,0 +1,42 @@ +<?php + +/** + * Generic schema interchange format that can be converted to a runtime + * representation (HTMLPurifier_ConfigSchema) or HTML documentation. Members + * are completely validated. + */ +class HTMLPurifier_ConfigSchema_Interchange +{ + + /** + * Name of the application this schema is describing. + */ + public $name; + + /** + * Array of Directive ID => array(directive info) + */ + public $directives = array(); + + /** + * Adds a directive array to $directives + */ + public function addDirective($directive) { + if (isset($this->directives[$i = $directive->id->toString()])) { + throw new HTMLPurifier_ConfigSchema_Exception("Cannot redefine directive '$i'"); + } + $this->directives[$i] = $directive; + } + + /** + * Convenience function to perform standard validation. Throws exception + * on failed validation. + */ + public function validate() { + $validator = new HTMLPurifier_ConfigSchema_Validator(); + return $validator->validate($this); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange/Directive.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange/Directive.php new file mode 100644 index 000000000..ac8be0d97 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange/Directive.php @@ -0,0 +1,77 @@ +<?php + +/** + * Interchange component class describing configuration directives. + */ +class HTMLPurifier_ConfigSchema_Interchange_Directive +{ + + /** + * ID of directive, instance of HTMLPurifier_ConfigSchema_Interchange_Id. + */ + public $id; + + /** + * String type, e.g. 'integer' or 'istring'. + */ + public $type; + + /** + * Default value, e.g. 3 or 'DefaultVal'. + */ + public $default; + + /** + * HTML description. + */ + public $description; + + /** + * Boolean whether or not null is allowed as a value. + */ + public $typeAllowsNull = false; + + /** + * Lookup table of allowed scalar values, e.g. array('allowed' => true). + * Null if all values are allowed. + */ + public $allowed; + + /** + * List of aliases for the directive, + * e.g. array(new HTMLPurifier_ConfigSchema_Interchange_Id('Ns', 'Dir'))). + */ + public $aliases = array(); + + /** + * Hash of value aliases, e.g. array('alt' => 'real'). Null if value + * aliasing is disabled (necessary for non-scalar types). + */ + public $valueAliases; + + /** + * Version of HTML Purifier the directive was introduced, e.g. '1.3.1'. + * Null if the directive has always existed. + */ + public $version; + + /** + * ID of directive that supercedes this old directive, is an instance + * of HTMLPurifier_ConfigSchema_Interchange_Id. Null if not deprecated. + */ + public $deprecatedUse; + + /** + * Version of HTML Purifier this directive was deprecated. Null if not + * deprecated. + */ + public $deprecatedVersion; + + /** + * List of external projects this directive depends on, e.g. array('CSSTidy'). + */ + public $external = array(); + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange/Id.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange/Id.php new file mode 100644 index 000000000..b9b3c6f5c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Interchange/Id.php @@ -0,0 +1,37 @@ +<?php + +/** + * Represents a directive ID in the interchange format. + */ +class HTMLPurifier_ConfigSchema_Interchange_Id +{ + + public $key; + + public function __construct($key) { + $this->key = $key; + } + + /** + * @warning This is NOT magic, to ensure that people don't abuse SPL and + * cause problems for PHP 5.0 support. + */ + public function toString() { + return $this->key; + } + + public function getRootNamespace() { + return substr($this->key, 0, strpos($this->key, ".")); + } + + public function getDirective() { + return substr($this->key, strpos($this->key, ".") + 1); + } + + public static function make($id) { + return new HTMLPurifier_ConfigSchema_Interchange_Id($id); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/InterchangeBuilder.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/InterchangeBuilder.php new file mode 100644 index 000000000..785b72ce8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/InterchangeBuilder.php @@ -0,0 +1,180 @@ +<?php + +class HTMLPurifier_ConfigSchema_InterchangeBuilder +{ + + /** + * Used for processing DEFAULT, nothing else. + */ + protected $varParser; + + public function __construct($varParser = null) { + $this->varParser = $varParser ? $varParser : new HTMLPurifier_VarParser_Native(); + } + + public static function buildFromDirectory($dir = null) { + $builder = new HTMLPurifier_ConfigSchema_InterchangeBuilder(); + $interchange = new HTMLPurifier_ConfigSchema_Interchange(); + return $builder->buildDir($interchange, $dir); + } + + public function buildDir($interchange, $dir = null) { + if (!$dir) $dir = HTMLPURIFIER_PREFIX . '/HTMLPurifier/ConfigSchema/schema'; + if (file_exists($dir . '/info.ini')) { + $info = parse_ini_file($dir . '/info.ini'); + $interchange->name = $info['name']; + } + + $files = array(); + $dh = opendir($dir); + while (false !== ($file = readdir($dh))) { + if (!$file || $file[0] == '.' || strrchr($file, '.') !== '.txt') { + continue; + } + $files[] = $file; + } + closedir($dh); + + sort($files); + foreach ($files as $file) { + $this->buildFile($interchange, $dir . '/' . $file); + } + + return $interchange; + } + + public function buildFile($interchange, $file) { + $parser = new HTMLPurifier_StringHashParser(); + $this->build( + $interchange, + new HTMLPurifier_StringHash( $parser->parseFile($file) ) + ); + } + + /** + * Builds an interchange object based on a hash. + * @param $interchange HTMLPurifier_ConfigSchema_Interchange object to build + * @param $hash HTMLPurifier_ConfigSchema_StringHash source data + */ + public function build($interchange, $hash) { + if (!$hash instanceof HTMLPurifier_StringHash) { + $hash = new HTMLPurifier_StringHash($hash); + } + if (!isset($hash['ID'])) { + throw new HTMLPurifier_ConfigSchema_Exception('Hash does not have any ID'); + } + if (strpos($hash['ID'], '.') === false) { + if (count($hash) == 2 && isset($hash['DESCRIPTION'])) { + $hash->offsetGet('DESCRIPTION'); // prevent complaining + } else { + throw new HTMLPurifier_ConfigSchema_Exception('All directives must have a namespace'); + } + } else { + $this->buildDirective($interchange, $hash); + } + $this->_findUnused($hash); + } + + public function buildDirective($interchange, $hash) { + $directive = new HTMLPurifier_ConfigSchema_Interchange_Directive(); + + // These are required elements: + $directive->id = $this->id($hash->offsetGet('ID')); + $id = $directive->id->toString(); // convenience + + if (isset($hash['TYPE'])) { + $type = explode('/', $hash->offsetGet('TYPE')); + if (isset($type[1])) $directive->typeAllowsNull = true; + $directive->type = $type[0]; + } else { + throw new HTMLPurifier_ConfigSchema_Exception("TYPE in directive hash '$id' not defined"); + } + + if (isset($hash['DEFAULT'])) { + try { + $directive->default = $this->varParser->parse($hash->offsetGet('DEFAULT'), $directive->type, $directive->typeAllowsNull); + } catch (HTMLPurifier_VarParserException $e) { + throw new HTMLPurifier_ConfigSchema_Exception($e->getMessage() . " in DEFAULT in directive hash '$id'"); + } + } + + if (isset($hash['DESCRIPTION'])) { + $directive->description = $hash->offsetGet('DESCRIPTION'); + } + + if (isset($hash['ALLOWED'])) { + $directive->allowed = $this->lookup($this->evalArray($hash->offsetGet('ALLOWED'))); + } + + if (isset($hash['VALUE-ALIASES'])) { + $directive->valueAliases = $this->evalArray($hash->offsetGet('VALUE-ALIASES')); + } + + if (isset($hash['ALIASES'])) { + $raw_aliases = trim($hash->offsetGet('ALIASES')); + $aliases = preg_split('/\s*,\s*/', $raw_aliases); + foreach ($aliases as $alias) { + $directive->aliases[] = $this->id($alias); + } + } + + if (isset($hash['VERSION'])) { + $directive->version = $hash->offsetGet('VERSION'); + } + + if (isset($hash['DEPRECATED-USE'])) { + $directive->deprecatedUse = $this->id($hash->offsetGet('DEPRECATED-USE')); + } + + if (isset($hash['DEPRECATED-VERSION'])) { + $directive->deprecatedVersion = $hash->offsetGet('DEPRECATED-VERSION'); + } + + if (isset($hash['EXTERNAL'])) { + $directive->external = preg_split('/\s*,\s*/', trim($hash->offsetGet('EXTERNAL'))); + } + + $interchange->addDirective($directive); + } + + /** + * Evaluates an array PHP code string without array() wrapper + */ + protected function evalArray($contents) { + return eval('return array('. $contents .');'); + } + + /** + * Converts an array list into a lookup array. + */ + protected function lookup($array) { + $ret = array(); + foreach ($array as $val) $ret[$val] = true; + return $ret; + } + + /** + * Convenience function that creates an HTMLPurifier_ConfigSchema_Interchange_Id + * object based on a string Id. + */ + protected function id($id) { + return HTMLPurifier_ConfigSchema_Interchange_Id::make($id); + } + + /** + * Triggers errors for any unused keys passed in the hash; such keys + * may indicate typos, missing values, etc. + * @param $hash Instance of ConfigSchema_StringHash to check. + */ + protected function _findUnused($hash) { + $accessed = $hash->getAccessed(); + foreach ($hash as $k => $v) { + if (!isset($accessed[$k])) { + trigger_error("String hash key '$k' not used by builder", E_USER_NOTICE); + } + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Validator.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Validator.php new file mode 100644 index 000000000..f374f6a02 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/Validator.php @@ -0,0 +1,206 @@ +<?php + +/** + * Performs validations on HTMLPurifier_ConfigSchema_Interchange + * + * @note If you see '// handled by InterchangeBuilder', that means a + * design decision in that class would prevent this validation from + * ever being necessary. We have them anyway, however, for + * redundancy. + */ +class HTMLPurifier_ConfigSchema_Validator +{ + + /** + * Easy to access global objects. + */ + protected $interchange, $aliases; + + /** + * Context-stack to provide easy to read error messages. + */ + protected $context = array(); + + /** + * HTMLPurifier_VarParser to test default's type. + */ + protected $parser; + + public function __construct() { + $this->parser = new HTMLPurifier_VarParser(); + } + + /** + * Validates a fully-formed interchange object. Throws an + * HTMLPurifier_ConfigSchema_Exception if there's a problem. + */ + public function validate($interchange) { + $this->interchange = $interchange; + $this->aliases = array(); + // PHP is a bit lax with integer <=> string conversions in + // arrays, so we don't use the identical !== comparison + foreach ($interchange->directives as $i => $directive) { + $id = $directive->id->toString(); + if ($i != $id) $this->error(false, "Integrity violation: key '$i' does not match internal id '$id'"); + $this->validateDirective($directive); + } + return true; + } + + /** + * Validates a HTMLPurifier_ConfigSchema_Interchange_Id object. + */ + public function validateId($id) { + $id_string = $id->toString(); + $this->context[] = "id '$id_string'"; + if (!$id instanceof HTMLPurifier_ConfigSchema_Interchange_Id) { + // handled by InterchangeBuilder + $this->error(false, 'is not an instance of HTMLPurifier_ConfigSchema_Interchange_Id'); + } + // keys are now unconstrained (we might want to narrow down to A-Za-z0-9.) + // we probably should check that it has at least one namespace + $this->with($id, 'key') + ->assertNotEmpty() + ->assertIsString(); // implicit assertIsString handled by InterchangeBuilder + array_pop($this->context); + } + + /** + * Validates a HTMLPurifier_ConfigSchema_Interchange_Directive object. + */ + public function validateDirective($d) { + $id = $d->id->toString(); + $this->context[] = "directive '$id'"; + $this->validateId($d->id); + + $this->with($d, 'description') + ->assertNotEmpty(); + + // BEGIN - handled by InterchangeBuilder + $this->with($d, 'type') + ->assertNotEmpty(); + $this->with($d, 'typeAllowsNull') + ->assertIsBool(); + try { + // This also tests validity of $d->type + $this->parser->parse($d->default, $d->type, $d->typeAllowsNull); + } catch (HTMLPurifier_VarParserException $e) { + $this->error('default', 'had error: ' . $e->getMessage()); + } + // END - handled by InterchangeBuilder + + if (!is_null($d->allowed) || !empty($d->valueAliases)) { + // allowed and valueAliases require that we be dealing with + // strings, so check for that early. + $d_int = HTMLPurifier_VarParser::$types[$d->type]; + if (!isset(HTMLPurifier_VarParser::$stringTypes[$d_int])) { + $this->error('type', 'must be a string type when used with allowed or value aliases'); + } + } + + $this->validateDirectiveAllowed($d); + $this->validateDirectiveValueAliases($d); + $this->validateDirectiveAliases($d); + + array_pop($this->context); + } + + /** + * Extra validation if $allowed member variable of + * HTMLPurifier_ConfigSchema_Interchange_Directive is defined. + */ + public function validateDirectiveAllowed($d) { + if (is_null($d->allowed)) return; + $this->with($d, 'allowed') + ->assertNotEmpty() + ->assertIsLookup(); // handled by InterchangeBuilder + if (is_string($d->default) && !isset($d->allowed[$d->default])) { + $this->error('default', 'must be an allowed value'); + } + $this->context[] = 'allowed'; + foreach ($d->allowed as $val => $x) { + if (!is_string($val)) $this->error("value $val", 'must be a string'); + } + array_pop($this->context); + } + + /** + * Extra validation if $valueAliases member variable of + * HTMLPurifier_ConfigSchema_Interchange_Directive is defined. + */ + public function validateDirectiveValueAliases($d) { + if (is_null($d->valueAliases)) return; + $this->with($d, 'valueAliases') + ->assertIsArray(); // handled by InterchangeBuilder + $this->context[] = 'valueAliases'; + foreach ($d->valueAliases as $alias => $real) { + if (!is_string($alias)) $this->error("alias $alias", 'must be a string'); + if (!is_string($real)) $this->error("alias target $real from alias '$alias'", 'must be a string'); + if ($alias === $real) { + $this->error("alias '$alias'", "must not be an alias to itself"); + } + } + if (!is_null($d->allowed)) { + foreach ($d->valueAliases as $alias => $real) { + if (isset($d->allowed[$alias])) { + $this->error("alias '$alias'", 'must not be an allowed value'); + } elseif (!isset($d->allowed[$real])) { + $this->error("alias '$alias'", 'must be an alias to an allowed value'); + } + } + } + array_pop($this->context); + } + + /** + * Extra validation if $aliases member variable of + * HTMLPurifier_ConfigSchema_Interchange_Directive is defined. + */ + public function validateDirectiveAliases($d) { + $this->with($d, 'aliases') + ->assertIsArray(); // handled by InterchangeBuilder + $this->context[] = 'aliases'; + foreach ($d->aliases as $alias) { + $this->validateId($alias); + $s = $alias->toString(); + if (isset($this->interchange->directives[$s])) { + $this->error("alias '$s'", 'collides with another directive'); + } + if (isset($this->aliases[$s])) { + $other_directive = $this->aliases[$s]; + $this->error("alias '$s'", "collides with alias for directive '$other_directive'"); + } + $this->aliases[$s] = $d->id->toString(); + } + array_pop($this->context); + } + + // protected helper functions + + /** + * Convenience function for generating HTMLPurifier_ConfigSchema_ValidatorAtom + * for validating simple member variables of objects. + */ + protected function with($obj, $member) { + return new HTMLPurifier_ConfigSchema_ValidatorAtom($this->getFormattedContext(), $obj, $member); + } + + /** + * Emits an error, providing helpful context. + */ + protected function error($target, $msg) { + if ($target !== false) $prefix = ucfirst($target) . ' in ' . $this->getFormattedContext(); + else $prefix = ucfirst($this->getFormattedContext()); + throw new HTMLPurifier_ConfigSchema_Exception(trim($prefix . ' ' . $msg)); + } + + /** + * Returns a formatted context string. + */ + protected function getFormattedContext() { + return implode(' in ', array_reverse($this->context)); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/ValidatorAtom.php b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/ValidatorAtom.php new file mode 100644 index 000000000..b95aea18c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/ValidatorAtom.php @@ -0,0 +1,66 @@ +<?php + +/** + * Fluent interface for validating the contents of member variables. + * This should be immutable. See HTMLPurifier_ConfigSchema_Validator for + * use-cases. We name this an 'atom' because it's ONLY for validations that + * are independent and usually scalar. + */ +class HTMLPurifier_ConfigSchema_ValidatorAtom +{ + + protected $context, $obj, $member, $contents; + + public function __construct($context, $obj, $member) { + $this->context = $context; + $this->obj = $obj; + $this->member = $member; + $this->contents =& $obj->$member; + } + + public function assertIsString() { + if (!is_string($this->contents)) $this->error('must be a string'); + return $this; + } + + public function assertIsBool() { + if (!is_bool($this->contents)) $this->error('must be a boolean'); + return $this; + } + + public function assertIsArray() { + if (!is_array($this->contents)) $this->error('must be an array'); + return $this; + } + + public function assertNotNull() { + if ($this->contents === null) $this->error('must not be null'); + return $this; + } + + public function assertAlnum() { + $this->assertIsString(); + if (!ctype_alnum($this->contents)) $this->error('must be alphanumeric'); + return $this; + } + + public function assertNotEmpty() { + if (empty($this->contents)) $this->error('must not be empty'); + return $this; + } + + public function assertIsLookup() { + $this->assertIsArray(); + foreach ($this->contents as $v) { + if ($v !== true) $this->error('must be a lookup array'); + } + return $this; + } + + protected function error($msg) { + throw new HTMLPurifier_ConfigSchema_Exception(ucfirst($this->member) . ' in ' . $this->context . ' ' . $msg); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema.ser b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema.ser Binary files differnew file mode 100644 index 000000000..b106bcf79 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema.ser diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedClasses.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedClasses.txt new file mode 100644 index 000000000..0517fed0a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedClasses.txt @@ -0,0 +1,8 @@ +Attr.AllowedClasses +TYPE: lookup/null +VERSION: 4.0.0 +DEFAULT: null +--DESCRIPTION-- +List of allowed class values in the class attribute. By default, this is null, +which means all classes are allowed. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedFrameTargets.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedFrameTargets.txt new file mode 100644 index 000000000..249edd647 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedFrameTargets.txt @@ -0,0 +1,12 @@ +Attr.AllowedFrameTargets +TYPE: lookup +DEFAULT: array() +--DESCRIPTION-- +Lookup table of all allowed link frame targets. Some commonly used link +targets include _blank, _self, _parent and _top. Values should be +lowercase, as validation will be done in a case-sensitive manner despite +W3C's recommendation. XHTML 1.0 Strict does not permit the target attribute +so this directive will have no effect in that doctype. XHTML 1.1 does not +enable the Target module by default, you will have to manually enable it +(see the module documentation for more details.) +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedRel.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedRel.txt new file mode 100644 index 000000000..9a8fa6a2e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedRel.txt @@ -0,0 +1,9 @@ +Attr.AllowedRel +TYPE: lookup +VERSION: 1.6.0 +DEFAULT: array() +--DESCRIPTION-- +List of allowed forward document relationships in the rel attribute. Common +values may be nofollow or print. By default, this is empty, meaning that no +document relationships are allowed. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedRev.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedRev.txt new file mode 100644 index 000000000..b01788348 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.AllowedRev.txt @@ -0,0 +1,9 @@ +Attr.AllowedRev +TYPE: lookup +VERSION: 1.6.0 +DEFAULT: array() +--DESCRIPTION-- +List of allowed reverse document relationships in the rev attribute. This +attribute is a bit of an edge-case; if you don't know what it is for, stay +away. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.ClassUseCDATA.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.ClassUseCDATA.txt new file mode 100644 index 000000000..e774b823b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.ClassUseCDATA.txt @@ -0,0 +1,19 @@ +Attr.ClassUseCDATA +TYPE: bool/null +DEFAULT: null +VERSION: 4.0.0 +--DESCRIPTION-- +If null, class will auto-detect the doctype and, if matching XHTML 1.1 or +XHTML 2.0, will use the restrictive NMTOKENS specification of class. Otherwise, +it will use a relaxed CDATA definition. If true, the relaxed CDATA definition +is forced; if false, the NMTOKENS definition is forced. To get behavior +of HTML Purifier prior to 4.0.0, set this directive to false. + +Some rational behind the auto-detection: +in previous versions of HTML Purifier, it was assumed that the form of +class was NMTOKENS, as specified by the XHTML Modularization (representing +XHTML 1.1 and XHTML 2.0). The DTDs for HTML 4.01 and XHTML 1.0, however +specify class as CDATA. HTML 5 effectively defines it as CDATA, but +with the additional constraint that each name should be unique (this is not +explicitly outlined in previous specifications). +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultImageAlt.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultImageAlt.txt new file mode 100644 index 000000000..533165e17 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultImageAlt.txt @@ -0,0 +1,11 @@ +Attr.DefaultImageAlt +TYPE: string/null +DEFAULT: null +VERSION: 3.2.0 +--DESCRIPTION-- +This is the content of the alt tag of an image if the user had not +previously specified an alt attribute. This applies to all images without +a valid alt attribute, as opposed to %Attr.DefaultInvalidImageAlt, which +only applies to invalid images, and overrides in the case of an invalid image. +Default behavior with null is to use the basename of the src tag for the alt. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultInvalidImage.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultInvalidImage.txt new file mode 100644 index 000000000..9eb7e3846 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultInvalidImage.txt @@ -0,0 +1,9 @@ +Attr.DefaultInvalidImage +TYPE: string +DEFAULT: '' +--DESCRIPTION-- +This is the default image an img tag will be pointed to if it does not have +a valid src attribute. In future versions, we may allow the image tag to +be removed completely, but due to design issues, this is not possible right +now. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultInvalidImageAlt.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultInvalidImageAlt.txt new file mode 100644 index 000000000..2f17bf477 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultInvalidImageAlt.txt @@ -0,0 +1,8 @@ +Attr.DefaultInvalidImageAlt +TYPE: string +DEFAULT: 'Invalid image' +--DESCRIPTION-- +This is the content of the alt tag of an invalid image if the user had not +previously specified an alt attribute. It has no effect when the image is +valid but there was no alt attribute present. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultTextDir.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultTextDir.txt new file mode 100644 index 000000000..52654b53a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.DefaultTextDir.txt @@ -0,0 +1,10 @@ +Attr.DefaultTextDir +TYPE: string +DEFAULT: 'ltr' +--DESCRIPTION-- +Defines the default text direction (ltr or rtl) of the document being +parsed. This generally is the same as the value of the dir attribute in +HTML, or ltr if that is not specified. +--ALLOWED-- +'ltr', 'rtl' +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.EnableID.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.EnableID.txt new file mode 100644 index 000000000..6440d2103 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.EnableID.txt @@ -0,0 +1,16 @@ +Attr.EnableID +TYPE: bool +DEFAULT: false +VERSION: 1.2.0 +--DESCRIPTION-- +Allows the ID attribute in HTML. This is disabled by default due to the +fact that without proper configuration user input can easily break the +validation of a webpage by specifying an ID that is already on the +surrounding HTML. If you don't mind throwing caution to the wind, enable +this directive, but I strongly recommend you also consider blacklisting IDs +you use (%Attr.IDBlacklist) or prefixing all user supplied IDs +(%Attr.IDPrefix). When set to true HTML Purifier reverts to the behavior of +pre-1.2.0 versions. +--ALIASES-- +HTML.EnableAttrID +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.ForbiddenClasses.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.ForbiddenClasses.txt new file mode 100644 index 000000000..f31d226f5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.ForbiddenClasses.txt @@ -0,0 +1,8 @@ +Attr.ForbiddenClasses +TYPE: lookup +VERSION: 4.0.0 +DEFAULT: array() +--DESCRIPTION-- +List of forbidden class values in the class attribute. By default, this is +empty, which means that no classes are forbidden. See also %Attr.AllowedClasses. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDBlacklist.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDBlacklist.txt new file mode 100644 index 000000000..5f2b5e3d2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDBlacklist.txt @@ -0,0 +1,5 @@ +Attr.IDBlacklist +TYPE: list +DEFAULT: array() +DESCRIPTION: Array of IDs not allowed in the document. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDBlacklistRegexp.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDBlacklistRegexp.txt new file mode 100644 index 000000000..6f5824586 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDBlacklistRegexp.txt @@ -0,0 +1,9 @@ +Attr.IDBlacklistRegexp +TYPE: string/null +VERSION: 1.6.0 +DEFAULT: NULL +--DESCRIPTION-- +PCRE regular expression to be matched against all IDs. If the expression is +matches, the ID is rejected. Use this with care: may cause significant +degradation. ID matching is done after all other validation. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDPrefix.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDPrefix.txt new file mode 100644 index 000000000..cc49d43fd --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDPrefix.txt @@ -0,0 +1,12 @@ +Attr.IDPrefix +TYPE: string +VERSION: 1.2.0 +DEFAULT: '' +--DESCRIPTION-- +String to prefix to IDs. If you have no idea what IDs your pages may use, +you may opt to simply add a prefix to all user-submitted ID attributes so +that they are still usable, but will not conflict with core page IDs. +Example: setting the directive to 'user_' will result in a user submitted +'foo' to become 'user_foo' Be sure to set %HTML.EnableAttrID to true +before using this. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDPrefixLocal.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDPrefixLocal.txt new file mode 100644 index 000000000..2c5924a7a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Attr.IDPrefixLocal.txt @@ -0,0 +1,14 @@ +Attr.IDPrefixLocal +TYPE: string +VERSION: 1.2.0 +DEFAULT: '' +--DESCRIPTION-- +Temporary prefix for IDs used in conjunction with %Attr.IDPrefix. If you +need to allow multiple sets of user content on web page, you may need to +have a seperate prefix that changes with each iteration. This way, +seperately submitted user content displayed on the same page doesn't +clobber each other. Ideal values are unique identifiers for the content it +represents (i.e. the id of the row in the database). Be sure to add a +seperator (like an underscore) at the end. Warning: this directive will +not work unless %Attr.IDPrefix is set to a non-empty value! +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.AutoParagraph.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.AutoParagraph.txt new file mode 100644 index 000000000..d5caa1bb9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.AutoParagraph.txt @@ -0,0 +1,31 @@ +AutoFormat.AutoParagraph +TYPE: bool +VERSION: 2.0.1 +DEFAULT: false +--DESCRIPTION-- + +<p> + This directive turns on auto-paragraphing, where double newlines are + converted in to paragraphs whenever possible. Auto-paragraphing: +</p> +<ul> + <li>Always applies to inline elements or text in the root node,</li> + <li>Applies to inline elements or text with double newlines in nodes + that allow paragraph tags,</li> + <li>Applies to double newlines in paragraph tags</li> +</ul> +<p> + <code>p</code> tags must be allowed for this directive to take effect. + We do not use <code>br</code> tags for paragraphing, as that is + semantically incorrect. +</p> +<p> + To prevent auto-paragraphing as a content-producer, refrain from using + double-newlines except to specify a new paragraph or in contexts where + it has special meaning (whitespace usually has no meaning except in + tags like <code>pre</code>, so this should not be difficult.) To prevent + the paragraphing of inline text adjacent to block elements, wrap them + in <code>div</code> tags (the behavior is slightly different outside of + the root node.) +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.Custom.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.Custom.txt new file mode 100644 index 000000000..2a476481a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.Custom.txt @@ -0,0 +1,12 @@ +AutoFormat.Custom +TYPE: list +VERSION: 2.0.1 +DEFAULT: array() +--DESCRIPTION-- + +<p> + This directive can be used to add custom auto-format injectors. + Specify an array of injector names (class name minus the prefix) + or concrete implementations. Injector class must exist. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.DisplayLinkURI.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.DisplayLinkURI.txt new file mode 100644 index 000000000..663064a34 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.DisplayLinkURI.txt @@ -0,0 +1,11 @@ +AutoFormat.DisplayLinkURI +TYPE: bool +VERSION: 3.2.0 +DEFAULT: false +--DESCRIPTION-- +<p> + This directive turns on the in-text display of URIs in <a> tags, and disables + those links. For example, <a href="http://example.com">example</a> becomes + example (<a>http://example.com</a>). +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.Linkify.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.Linkify.txt new file mode 100644 index 000000000..3a48ba960 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.Linkify.txt @@ -0,0 +1,12 @@ +AutoFormat.Linkify +TYPE: bool +VERSION: 2.0.1 +DEFAULT: false +--DESCRIPTION-- + +<p> + This directive turns on linkification, auto-linking http, ftp and + https URLs. <code>a</code> tags with the <code>href</code> attribute + must be allowed. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.PurifierLinkify.DocURL.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.PurifierLinkify.DocURL.txt new file mode 100644 index 000000000..db58b1346 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.PurifierLinkify.DocURL.txt @@ -0,0 +1,12 @@ +AutoFormat.PurifierLinkify.DocURL +TYPE: string +VERSION: 2.0.1 +DEFAULT: '#%s' +ALIASES: AutoFormatParam.PurifierLinkifyDocURL +--DESCRIPTION-- +<p> + Location of configuration documentation to link to, let %s substitute + into the configuration's namespace and directive names sans the percent + sign. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.PurifierLinkify.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.PurifierLinkify.txt new file mode 100644 index 000000000..7996488be --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.PurifierLinkify.txt @@ -0,0 +1,12 @@ +AutoFormat.PurifierLinkify +TYPE: bool +VERSION: 2.0.1 +DEFAULT: false +--DESCRIPTION-- + +<p> + Internal auto-formatter that converts configuration directives in + syntax <a>%Namespace.Directive</a> to links. <code>a</code> tags + with the <code>href</code> attribute must be allowed. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions.txt new file mode 100644 index 000000000..35c393b4e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions.txt @@ -0,0 +1,11 @@ +AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions +TYPE: lookup +VERSION: 4.0.0 +DEFAULT: array('td' => true, 'th' => true) +--DESCRIPTION-- +<p> + When %AutoFormat.RemoveEmpty and %AutoFormat.RemoveEmpty.RemoveNbsp + are enabled, this directive defines what HTML elements should not be + removede if they have only a non-breaking space in them. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.RemoveNbsp.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.RemoveNbsp.txt new file mode 100644 index 000000000..ca17eb1dc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.RemoveNbsp.txt @@ -0,0 +1,15 @@ +AutoFormat.RemoveEmpty.RemoveNbsp +TYPE: bool +VERSION: 4.0.0 +DEFAULT: false +--DESCRIPTION-- +<p> + When enabled, HTML Purifier will treat any elements that contain only + non-breaking spaces as well as regular whitespace as empty, and remove + them when %AutoForamt.RemoveEmpty is enabled. +</p> +<p> + See %AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions for a list of elements + that don't have this behavior applied to them. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.txt new file mode 100644 index 000000000..34657ba47 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveEmpty.txt @@ -0,0 +1,46 @@ +AutoFormat.RemoveEmpty +TYPE: bool +VERSION: 3.2.0 +DEFAULT: false +--DESCRIPTION-- +<p> + When enabled, HTML Purifier will attempt to remove empty elements that + contribute no semantic information to the document. The following types + of nodes will be removed: +</p> +<ul><li> + Tags with no attributes and no content, and that are not empty + elements (remove <code><a></a></code> but not + <code><br /></code>), and + </li> + <li> + Tags with no content, except for:<ul> + <li>The <code>colgroup</code> element, or</li> + <li> + Elements with the <code>id</code> or <code>name</code> attribute, + when those attributes are permitted on those elements. + </li> + </ul></li> +</ul> +<p> + Please be very careful when using this functionality; while it may not + seem that empty elements contain useful information, they can alter the + layout of a document given appropriate styling. This directive is most + useful when you are processing machine-generated HTML, please avoid using + it on regular user HTML. +</p> +<p> + Elements that contain only whitespace will be treated as empty. Non-breaking + spaces, however, do not count as whitespace. See + %AutoFormat.RemoveEmpty.RemoveNbsp for alternate behavior. +</p> +<p> + This algorithm is not perfect; you may still notice some empty tags, + particularly if a node had elements, but those elements were later removed + because they were not permitted in that context, or tags that, after + being auto-closed by another tag, where empty. This is for safety reasons + to prevent clever code from breaking validation. The general rule of thumb: + if a tag looked empty on the way in, it will get removed; if HTML Purifier + made it empty, it will stay. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveSpansWithoutAttributes.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveSpansWithoutAttributes.txt new file mode 100644 index 000000000..dde990ab2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/AutoFormat.RemoveSpansWithoutAttributes.txt @@ -0,0 +1,11 @@ +AutoFormat.RemoveSpansWithoutAttributes +TYPE: bool +VERSION: 4.0.1 +DEFAULT: false +--DESCRIPTION-- +<p> + This directive causes <code>span</code> tags without any attributes + to be removed. It will also remove spans that had all attributes + removed during processing. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowImportant.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowImportant.txt new file mode 100644 index 000000000..b324608f7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowImportant.txt @@ -0,0 +1,8 @@ +CSS.AllowImportant +TYPE: bool +DEFAULT: false +VERSION: 3.1.0 +--DESCRIPTION-- +This parameter determines whether or not !important cascade modifiers should +be allowed in user CSS. If false, !important will stripped. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowTricky.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowTricky.txt new file mode 100644 index 000000000..748be0eec --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowTricky.txt @@ -0,0 +1,11 @@ +CSS.AllowTricky +TYPE: bool +DEFAULT: false +VERSION: 3.1.0 +--DESCRIPTION-- +This parameter determines whether or not to allow "tricky" CSS properties and +values. Tricky CSS properties/values can drastically modify page layout or +be used for deceptive practices but do not directly constitute a security risk. +For example, <code>display:none;</code> is considered a tricky property that +will only be allowed if this directive is set to true. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowedFonts.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowedFonts.txt new file mode 100644 index 000000000..3fd465406 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowedFonts.txt @@ -0,0 +1,12 @@ +CSS.AllowedFonts +TYPE: lookup/null +VERSION: 4.3.0 +DEFAULT: NULL +--DESCRIPTION-- +<p> + Allows you to manually specify a set of allowed fonts. If + <code>NULL</code>, all fonts are allowed. This directive + affects generic names (serif, sans-serif, monospace, cursive, + fantasy) as well as specific font families. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowedProperties.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowedProperties.txt new file mode 100644 index 000000000..460112ebe --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.AllowedProperties.txt @@ -0,0 +1,18 @@ +CSS.AllowedProperties +TYPE: lookup/null +VERSION: 3.1.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + If HTML Purifier's style attributes set is unsatisfactory for your needs, + you can overload it with your own list of tags to allow. Note that this + method is subtractive: it does its job by taking away from HTML Purifier + usual feature set, so you cannot add an attribute that HTML Purifier never + supported in the first place. +</p> +<p> + <strong>Warning:</strong> If another directive conflicts with the + elements here, <em>that</em> directive will win and override. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.DefinitionRev.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.DefinitionRev.txt new file mode 100644 index 000000000..5cb7dda3b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.DefinitionRev.txt @@ -0,0 +1,11 @@ +CSS.DefinitionRev +TYPE: int +VERSION: 2.0.0 +DEFAULT: 1 +--DESCRIPTION-- + +<p> + Revision identifier for your custom definition. See + %HTML.DefinitionRev for details. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.ForbiddenProperties.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.ForbiddenProperties.txt new file mode 100644 index 000000000..f1f5c5f12 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.ForbiddenProperties.txt @@ -0,0 +1,13 @@ +CSS.ForbiddenProperties +TYPE: lookup +VERSION: 4.2.0 +DEFAULT: array() +--DESCRIPTION-- +<p> + This is the logical inverse of %CSS.AllowedProperties, and it will + override that directive or any other directive. If possible, + %CSS.AllowedProperties is recommended over this directive, + because it can sometimes be difficult to tell whether or not you've + forbidden all of the CSS properties you truly would like to disallow. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.MaxImgLength.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.MaxImgLength.txt new file mode 100644 index 000000000..7a3291470 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.MaxImgLength.txt @@ -0,0 +1,16 @@ +CSS.MaxImgLength +TYPE: string/null +DEFAULT: '1200px' +VERSION: 3.1.1 +--DESCRIPTION-- +<p> + This parameter sets the maximum allowed length on <code>img</code> tags, + effectively the <code>width</code> and <code>height</code> properties. + Only absolute units of measurement (in, pt, pc, mm, cm) and pixels (px) are allowed. This is + in place to prevent imagecrash attacks, disable with null at your own risk. + This directive is similar to %HTML.MaxImgLength, and both should be + concurrently edited, although there are + subtle differences in the input format (the CSS max is a number with + a unit). +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.Proprietary.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.Proprietary.txt new file mode 100644 index 000000000..148eedb8b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.Proprietary.txt @@ -0,0 +1,10 @@ +CSS.Proprietary +TYPE: bool +VERSION: 3.0.0 +DEFAULT: false +--DESCRIPTION-- + +<p> + Whether or not to allow safe, proprietary CSS values. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.Trusted.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.Trusted.txt new file mode 100644 index 000000000..e733a61e8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/CSS.Trusted.txt @@ -0,0 +1,9 @@ +CSS.Trusted +TYPE: bool +VERSION: 4.2.1 +DEFAULT: false +--DESCRIPTION-- +Indicates whether or not the user's CSS input is trusted or not. If the +input is trusted, a more expansive set of allowed properties. See +also %HTML.Trusted. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.DefinitionImpl.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.DefinitionImpl.txt new file mode 100644 index 000000000..c486724c8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.DefinitionImpl.txt @@ -0,0 +1,14 @@ +Cache.DefinitionImpl +TYPE: string/null +VERSION: 2.0.0 +DEFAULT: 'Serializer' +--DESCRIPTION-- + +This directive defines which method to use when caching definitions, +the complex data-type that makes HTML Purifier tick. Set to null +to disable caching (not recommended, as you will see a definite +performance degradation). + +--ALIASES-- +Core.DefinitionCache +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPath.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPath.txt new file mode 100644 index 000000000..54036507d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPath.txt @@ -0,0 +1,13 @@ +Cache.SerializerPath +TYPE: string/null +VERSION: 2.0.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + Absolute path with no trailing slash to store serialized definitions in. + Default is within the + HTML Purifier library inside DefinitionCache/Serializer. This + path must be writable by the webserver. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt new file mode 100644 index 000000000..b2b83d9ab --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Cache.SerializerPermissions.txt @@ -0,0 +1,11 @@ +Cache.SerializerPermissions +TYPE: int +VERSION: 4.3.0 +DEFAULT: 0755 +--DESCRIPTION-- + +<p> + Directory permissions of the files and directories created inside + the DefinitionCache/Serializer or other custom serializer path. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.AggressivelyFixLt.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.AggressivelyFixLt.txt new file mode 100644 index 000000000..568cbf3b3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.AggressivelyFixLt.txt @@ -0,0 +1,18 @@ +Core.AggressivelyFixLt +TYPE: bool +VERSION: 2.1.0 +DEFAULT: true +--DESCRIPTION-- +<p> + This directive enables aggressive pre-filter fixes HTML Purifier can + perform in order to ensure that open angled-brackets do not get killed + during parsing stage. Enabling this will result in two preg_replace_callback + calls and at least two preg_replace calls for every HTML document parsed; + if your users make very well-formed HTML, you can set this directive false. + This has no effect when DirectLex is used. +</p> +<p> + <strong>Notice:</strong> This directive's default turned from false to true + in HTML Purifier 3.2.0. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.CollectErrors.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.CollectErrors.txt new file mode 100644 index 000000000..d7317911f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.CollectErrors.txt @@ -0,0 +1,12 @@ +Core.CollectErrors +TYPE: bool +VERSION: 2.0.0 +DEFAULT: false +--DESCRIPTION-- + +Whether or not to collect errors found while filtering the document. This +is a useful way to give feedback to your users. <strong>Warning:</strong> +Currently this feature is very patchy and experimental, with lots of +possible error messages not yet implemented. It will not cause any +problems, but it may not help your users either. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.ColorKeywords.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.ColorKeywords.txt new file mode 100644 index 000000000..c572c14ec --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.ColorKeywords.txt @@ -0,0 +1,29 @@ +Core.ColorKeywords +TYPE: hash +VERSION: 2.0.0 +--DEFAULT-- +array ( + 'maroon' => '#800000', + 'red' => '#FF0000', + 'orange' => '#FFA500', + 'yellow' => '#FFFF00', + 'olive' => '#808000', + 'purple' => '#800080', + 'fuchsia' => '#FF00FF', + 'white' => '#FFFFFF', + 'lime' => '#00FF00', + 'green' => '#008000', + 'navy' => '#000080', + 'blue' => '#0000FF', + 'aqua' => '#00FFFF', + 'teal' => '#008080', + 'black' => '#000000', + 'silver' => '#C0C0C0', + 'gray' => '#808080', +) +--DESCRIPTION-- + +Lookup array of color names to six digit hexadecimal number corresponding +to color, with preceding hash mark. Used when parsing colors. The lookup +is done in a case-insensitive manner. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.ConvertDocumentToFragment.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.ConvertDocumentToFragment.txt new file mode 100644 index 000000000..64b114fce --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.ConvertDocumentToFragment.txt @@ -0,0 +1,14 @@ +Core.ConvertDocumentToFragment +TYPE: bool +DEFAULT: true +--DESCRIPTION-- + +This parameter determines whether or not the filter should convert +input that is a full document with html and body tags to a fragment +of just the contents of a body tag. This parameter is simply something +HTML Purifier can do during an edge-case: for most inputs, this +processing is not necessary. + +--ALIASES-- +Core.AcceptFullDocuments +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.DirectLexLineNumberSyncInterval.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.DirectLexLineNumberSyncInterval.txt new file mode 100644 index 000000000..36f16e07e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.DirectLexLineNumberSyncInterval.txt @@ -0,0 +1,17 @@ +Core.DirectLexLineNumberSyncInterval +TYPE: int +VERSION: 2.0.0 +DEFAULT: 0 +--DESCRIPTION-- + +<p> + Specifies the number of tokens the DirectLex line number tracking + implementations should process before attempting to resyncronize the + current line count by manually counting all previous new-lines. When + at 0, this functionality is disabled. Lower values will decrease + performance, and this is only strictly necessary if the counting + algorithm is buggy (in which case you should report it as a bug). + This has no effect when %Core.MaintainLineNumbers is disabled or DirectLex is + not being used. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EnableIDNA.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EnableIDNA.txt new file mode 100644 index 000000000..ce243c35d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EnableIDNA.txt @@ -0,0 +1,9 @@ +Core.EnableIDNA +TYPE: bool +DEFAULT: false +VERSION: 4.4.0 +--DESCRIPTION-- +Allows international domain names in URLs. This configuration option +requires the PEAR Net_IDNA2 module to be installed. It operates by +punycoding any internationalized host names for maximum portability. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.Encoding.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.Encoding.txt new file mode 100644 index 000000000..8bfb47c3a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.Encoding.txt @@ -0,0 +1,15 @@ +Core.Encoding +TYPE: istring +DEFAULT: 'utf-8' +--DESCRIPTION-- +If for some reason you are unable to convert all webpages to UTF-8, you can +use this directive as a stop-gap compatibility change to let HTML Purifier +deal with non UTF-8 input. This technique has notable deficiencies: +absolutely no characters outside of the selected character encoding will be +preserved, not even the ones that have been ampersand escaped (this is due +to a UTF-8 specific <em>feature</em> that automatically resolves all +entities), making it pretty useless for anything except the most I18N-blind +applications, although %Core.EscapeNonASCIICharacters offers fixes this +trouble with another tradeoff. This directive only accepts ISO-8859-1 if +iconv is not enabled. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeInvalidChildren.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeInvalidChildren.txt new file mode 100644 index 000000000..4d5b5055c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeInvalidChildren.txt @@ -0,0 +1,10 @@ +Core.EscapeInvalidChildren +TYPE: bool +DEFAULT: false +--DESCRIPTION-- +When true, a child is found that is not allowed in the context of the +parent element will be transformed into text as if it were ASCII. When +false, that element and all internal tags will be dropped, though text will +be preserved. There is no option for dropping the element but preserving +child nodes. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeInvalidTags.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeInvalidTags.txt new file mode 100644 index 000000000..a7a5b249b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeInvalidTags.txt @@ -0,0 +1,7 @@ +Core.EscapeInvalidTags +TYPE: bool +DEFAULT: false +--DESCRIPTION-- +When true, invalid tags will be written back to the document as plain text. +Otherwise, they are silently dropped. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeNonASCIICharacters.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeNonASCIICharacters.txt new file mode 100644 index 000000000..abb499948 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.EscapeNonASCIICharacters.txt @@ -0,0 +1,13 @@ +Core.EscapeNonASCIICharacters +TYPE: bool +VERSION: 1.4.0 +DEFAULT: false +--DESCRIPTION-- +This directive overcomes a deficiency in %Core.Encoding by blindly +converting all non-ASCII characters into decimal numeric entities before +converting it to its native encoding. This means that even characters that +can be expressed in the non-UTF-8 encoding will be entity-ized, which can +be a real downer for encodings like Big5. It also assumes that the ASCII +repetoire is available, although this is the case for almost all encodings. +Anyway, use UTF-8! +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.HiddenElements.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.HiddenElements.txt new file mode 100644 index 000000000..915391edb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.HiddenElements.txt @@ -0,0 +1,19 @@ +Core.HiddenElements +TYPE: lookup +--DEFAULT-- +array ( + 'script' => true, + 'style' => true, +) +--DESCRIPTION-- + +<p> + This directive is a lookup array of elements which should have their + contents removed when they are not allowed by the HTML definition. + For example, the contents of a <code>script</code> tag are not + normally shown in a document, so if script tags are to be removed, + their contents should be removed to. This is opposed to a <code>b</code> + tag, which defines some presentational changes but does not hide its + contents. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.Language.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.Language.txt new file mode 100644 index 000000000..233fca14f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.Language.txt @@ -0,0 +1,10 @@ +Core.Language +TYPE: string +VERSION: 2.0.0 +DEFAULT: 'en' +--DESCRIPTION-- + +ISO 639 language code for localizable things in HTML Purifier to use, +which is mainly error reporting. There is currently only an English (en) +translation, so this directive is currently useless. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.LexerImpl.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.LexerImpl.txt new file mode 100644 index 000000000..8983e2cca --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.LexerImpl.txt @@ -0,0 +1,34 @@ +Core.LexerImpl +TYPE: mixed/null +VERSION: 2.0.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + This parameter determines what lexer implementation can be used. The + valid values are: +</p> +<dl> + <dt><em>null</em></dt> + <dd> + Recommended, the lexer implementation will be auto-detected based on + your PHP-version and configuration. + </dd> + <dt><em>string</em> lexer identifier</dt> + <dd> + This is a slim way of manually overridding the implementation. + Currently recognized values are: DOMLex (the default PHP5 +implementation) + and DirectLex (the default PHP4 implementation). Only use this if + you know what you are doing: usually, the auto-detection will + manage things for cases you aren't even aware of. + </dd> + <dt><em>object</em> lexer instance</dt> + <dd> + Super-advanced: you can specify your own, custom, implementation that + implements the interface defined by <code>HTMLPurifier_Lexer</code>. + I may remove this option simply because I don't expect anyone + to use it. + </dd> +</dl> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.MaintainLineNumbers.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.MaintainLineNumbers.txt new file mode 100644 index 000000000..eb841a759 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.MaintainLineNumbers.txt @@ -0,0 +1,16 @@ +Core.MaintainLineNumbers +TYPE: bool/null +VERSION: 2.0.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + If true, HTML Purifier will add line number information to all tokens. + This is useful when error reporting is turned on, but can result in + significant performance degradation and should not be used when + unnecessary. This directive must be used with the DirectLex lexer, + as the DOMLex lexer does not (yet) support this functionality. + If the value is null, an appropriate value will be selected based + on other configuration. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.NormalizeNewlines.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.NormalizeNewlines.txt new file mode 100644 index 000000000..d77f5360d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.NormalizeNewlines.txt @@ -0,0 +1,11 @@ +Core.NormalizeNewlines +TYPE: bool +VERSION: 4.2.0 +DEFAULT: true +--DESCRIPTION-- +<p> + Whether or not to normalize newlines to the operating + system default. When <code>false</code>, HTML Purifier + will attempt to preserve mixed newline files. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveInvalidImg.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveInvalidImg.txt new file mode 100644 index 000000000..4070c2a0d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveInvalidImg.txt @@ -0,0 +1,12 @@ +Core.RemoveInvalidImg +TYPE: bool +DEFAULT: true +VERSION: 1.3.0 +--DESCRIPTION-- + +<p> + This directive enables pre-emptive URI checking in <code>img</code> + tags, as the attribute validation strategy is not authorized to + remove elements from the document. Revert to pre-1.3.0 behavior by setting to false. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveProcessingInstructions.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveProcessingInstructions.txt new file mode 100644 index 000000000..3397d9f71 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveProcessingInstructions.txt @@ -0,0 +1,11 @@ +Core.RemoveProcessingInstructions +TYPE: bool +VERSION: 4.2.0 +DEFAULT: false +--DESCRIPTION-- +Instead of escaping processing instructions in the form <code><? ... +?></code>, remove it out-right. This may be useful if the HTML +you are validating contains XML processing instruction gunk, however, +it can also be user-unfriendly for people attempting to post PHP +snippets. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveScriptContents.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveScriptContents.txt new file mode 100644 index 000000000..a4cd966df --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Core.RemoveScriptContents.txt @@ -0,0 +1,12 @@ +Core.RemoveScriptContents +TYPE: bool/null +DEFAULT: NULL +VERSION: 2.0.0 +DEPRECATED-VERSION: 2.1.0 +DEPRECATED-USE: Core.HiddenElements +--DESCRIPTION-- +<p> + This directive enables HTML Purifier to remove not only script tags + but all of their contents. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.Custom.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.Custom.txt new file mode 100644 index 000000000..3db50ef20 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.Custom.txt @@ -0,0 +1,11 @@ +Filter.Custom +TYPE: list +VERSION: 3.1.0 +DEFAULT: array() +--DESCRIPTION-- +<p> + This directive can be used to add custom filters; it is nearly the + equivalent of the now deprecated <code>HTMLPurifier->addFilter()</code> + method. Specify an array of concrete implementations. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.Escaping.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.Escaping.txt new file mode 100644 index 000000000..16829bcda --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.Escaping.txt @@ -0,0 +1,14 @@ +Filter.ExtractStyleBlocks.Escaping +TYPE: bool +VERSION: 3.0.0 +DEFAULT: true +ALIASES: Filter.ExtractStyleBlocksEscaping, FilterParam.ExtractStyleBlocksEscaping +--DESCRIPTION-- + +<p> + Whether or not to escape the dangerous characters <, > and & + as \3C, \3E and \26, respectively. This is can be safely set to false + if the contents of StyleBlocks will be placed in an external stylesheet, + where there is no risk of it being interpreted as HTML. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.Scope.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.Scope.txt new file mode 100644 index 000000000..7f95f54d1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.Scope.txt @@ -0,0 +1,29 @@ +Filter.ExtractStyleBlocks.Scope +TYPE: string/null +VERSION: 3.0.0 +DEFAULT: NULL +ALIASES: Filter.ExtractStyleBlocksScope, FilterParam.ExtractStyleBlocksScope +--DESCRIPTION-- + +<p> + If you would like users to be able to define external stylesheets, but + only allow them to specify CSS declarations for a specific node and + prevent them from fiddling with other elements, use this directive. + It accepts any valid CSS selector, and will prepend this to any + CSS declaration extracted from the document. For example, if this + directive is set to <code>#user-content</code> and a user uses the + selector <code>a:hover</code>, the final selector will be + <code>#user-content a:hover</code>. +</p> +<p> + The comma shorthand may be used; consider the above example, with + <code>#user-content, #user-content2</code>, the final selector will + be <code>#user-content a:hover, #user-content2 a:hover</code>. +</p> +<p> + <strong>Warning:</strong> It is possible for users to bypass this measure + using a naughty + selector. This is a bug in CSS Tidy 1.3, not HTML + Purifier, and I am working to get it fixed. Until then, HTML Purifier + performs a basic check to prevent this. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.TidyImpl.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.TidyImpl.txt new file mode 100644 index 000000000..6c231b2d7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.TidyImpl.txt @@ -0,0 +1,16 @@ +Filter.ExtractStyleBlocks.TidyImpl +TYPE: mixed/null +VERSION: 3.1.0 +DEFAULT: NULL +ALIASES: FilterParam.ExtractStyleBlocksTidyImpl +--DESCRIPTION-- +<p> + If left NULL, HTML Purifier will attempt to instantiate a <code>csstidy</code> + class to use for internal cleaning. This will usually be good enough. +</p> +<p> + However, for trusted user input, you can set this to <code>false</code> to + disable cleaning. In addition, you can supply your own concrete implementation + of Tidy's interface to use, although I don't know why you'd want to do that. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.txt new file mode 100644 index 000000000..078d08741 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.ExtractStyleBlocks.txt @@ -0,0 +1,74 @@ +Filter.ExtractStyleBlocks +TYPE: bool +VERSION: 3.1.0 +DEFAULT: false +EXTERNAL: CSSTidy +--DESCRIPTION-- +<p> + This directive turns on the style block extraction filter, which removes + <code>style</code> blocks from input HTML, cleans them up with CSSTidy, + and places them in the <code>StyleBlocks</code> context variable, for further + use by you, usually to be placed in an external stylesheet, or a + <code>style</code> block in the <code>head</code> of your document. +</p> +<p> + Sample usage: +</p> +<pre><![CDATA[ +<?php + header('Content-type: text/html; charset=utf-8'); + echo '<?xml version="1.0" encoding="UTF-8"?>'; +?> +<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" + "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd"> +<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"> +<head> + <title>Filter.ExtractStyleBlocks</title> +<?php + require_once '/path/to/library/HTMLPurifier.auto.php'; + require_once '/path/to/csstidy.class.php'; + + $dirty = '<style>body {color:#F00;}</style> Some text'; + + $config = HTMLPurifier_Config::createDefault(); + $config->set('Filter', 'ExtractStyleBlocks', true); + $purifier = new HTMLPurifier($config); + + $html = $purifier->purify($dirty); + + // This implementation writes the stylesheets to the styles/ directory. + // You can also echo the styles inside the document, but it's a bit + // more difficult to make sure they get interpreted properly by + // browsers; try the usual CSS armoring techniques. + $styles = $purifier->context->get('StyleBlocks'); + $dir = 'styles/'; + if (!is_dir($dir)) mkdir($dir); + $hash = sha1($_GET['html']); + foreach ($styles as $i => $style) { + file_put_contents($name = $dir . $hash . "_$i"); + echo '<link rel="stylesheet" type="text/css" href="'.$name.'" />'; + } +?> +</head> +<body> + <div> + <?php echo $html; ?> + </div> +</b]]><![CDATA[ody> +</html> +]]></pre> +<p> + <strong>Warning:</strong> It is possible for a user to mount an + imagecrash attack using this CSS. Counter-measures are difficult; + it is not simply enough to limit the range of CSS lengths (using + relative lengths with many nesting levels allows for large values + to be attained without actually specifying them in the stylesheet), + and the flexible nature of selectors makes it difficult to selectively + disable lengths on image tags (HTML Purifier, however, does disable + CSS width and height in inline styling). There are probably two effective + counter measures: an explicit width and height set to auto in all + images in your document (unlikely) or the disabling of width and + height (somewhat reasonable). Whether or not these measures should be + used is left to the reader. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.YouTube.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.YouTube.txt new file mode 100644 index 000000000..321eaa2d8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Filter.YouTube.txt @@ -0,0 +1,16 @@ +Filter.YouTube +TYPE: bool +VERSION: 3.1.0 +DEFAULT: false +--DESCRIPTION-- +<p> + <strong>Warning:</strong> Deprecated in favor of %HTML.SafeObject and + %Output.FlashCompat (turn both on to allow YouTube videos and other + Flash content). +</p> +<p> + This directive enables YouTube video embedding in HTML Purifier. Check + <a href="http://htmlpurifier.org/docs/enduser-youtube.html">this document + on embedding videos</a> for more information on what this filter does. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Allowed.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Allowed.txt new file mode 100644 index 000000000..0b2c106da --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Allowed.txt @@ -0,0 +1,25 @@ +HTML.Allowed +TYPE: itext/null +VERSION: 2.0.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + This is a preferred convenience directive that combines + %HTML.AllowedElements and %HTML.AllowedAttributes. + Specify elements and attributes that are allowed using: + <code>element1[attr1|attr2],element2...</code>. For example, + if you would like to only allow paragraphs and links, specify + <code>a[href],p</code>. You can specify attributes that apply + to all elements using an asterisk, e.g. <code>*[lang]</code>. + You can also use newlines instead of commas to separate elements. +</p> +<p> + <strong>Warning</strong>: + All of the constraints on the component directives are still enforced. + The syntax is a <em>subset</em> of TinyMCE's <code>valid_elements</code> + whitelist: directly copy-pasting it here will probably result in + broken whitelists. If %HTML.AllowedElements or %HTML.AllowedAttributes + are set, this directive has no effect. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedAttributes.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedAttributes.txt new file mode 100644 index 000000000..fcf093f17 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedAttributes.txt @@ -0,0 +1,19 @@ +HTML.AllowedAttributes +TYPE: lookup/null +VERSION: 1.3.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + If HTML Purifier's attribute set is unsatisfactory, overload it! + The syntax is "tag.attr" or "*.attr" for the global attributes + (style, id, class, dir, lang, xml:lang). +</p> +<p> + <strong>Warning:</strong> If another directive conflicts with the + elements here, <em>that</em> directive will win and override. For + example, %HTML.EnableAttrID will take precedence over *.id in this + directive. You must set that directive to true before you can use + IDs at all. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedComments.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedComments.txt new file mode 100644 index 000000000..140e21423 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedComments.txt @@ -0,0 +1,10 @@ +HTML.AllowedComments +TYPE: lookup +VERSION: 4.4.0 +DEFAULT: array() +--DESCRIPTION-- +A whitelist which indicates what explicit comment bodies should be +allowed, modulo leading and trailing whitespace. See also %HTML.AllowedCommentsRegexp +(these directives are union'ed together, so a comment is considered +valid if any directive deems it valid.) +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedCommentsRegexp.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedCommentsRegexp.txt new file mode 100644 index 000000000..f22e977d4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedCommentsRegexp.txt @@ -0,0 +1,15 @@ +HTML.AllowedCommentsRegexp +TYPE: string/null +VERSION: 4.4.0 +DEFAULT: NULL +--DESCRIPTION-- +A regexp, which if it matches the body of a comment, indicates that +it should be allowed. Trailing and leading spaces are removed prior +to running this regular expression. +<strong>Warning:</strong> Make sure you specify +correct anchor metacharacters <code>^regex$</code>, otherwise you may accept +comments that you did not mean to! In particular, the regex <code>/foo|bar/</code> +is probably not sufficiently strict, since it also allows <code>foobar</code>. +See also %HTML.AllowedComments (these directives are union'ed together, +so a comment is considered valid if any directive deems it valid.) +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedElements.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedElements.txt new file mode 100644 index 000000000..1d3fa7907 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedElements.txt @@ -0,0 +1,23 @@ +HTML.AllowedElements +TYPE: lookup/null +VERSION: 1.3.0 +DEFAULT: NULL +--DESCRIPTION-- +<p> + If HTML Purifier's tag set is unsatisfactory for your needs, you can + overload it with your own list of tags to allow. If you change + this, you probably also want to change %HTML.AllowedAttributes; see + also %HTML.Allowed which lets you set allowed elements and + attributes at the same time. +</p> +<p> + If you attempt to allow an element that HTML Purifier does not know + about, HTML Purifier will raise an error. You will need to manually + tell HTML Purifier about this element by using the + <a href="http://htmlpurifier.org/docs/enduser-customize.html">advanced customization features.</a> +</p> +<p> + <strong>Warning:</strong> If another directive conflicts with the + elements here, <em>that</em> directive will win and override. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedModules.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedModules.txt new file mode 100644 index 000000000..5a59a55c0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.AllowedModules.txt @@ -0,0 +1,20 @@ +HTML.AllowedModules +TYPE: lookup/null +VERSION: 2.0.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + A doctype comes with a set of usual modules to use. Without having + to mucking about with the doctypes, you can quickly activate or + disable these modules by specifying which modules you wish to allow + with this directive. This is most useful for unit testing specific + modules, although end users may find it useful for their own ends. +</p> +<p> + If you specify a module that does not exist, the manager will silently + fail to use it, so be careful! User-defined modules are not affected + by this directive. Modules defined in %HTML.CoreModules are not + affected by this directive. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Attr.Name.UseCDATA.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Attr.Name.UseCDATA.txt new file mode 100644 index 000000000..151fb7b82 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Attr.Name.UseCDATA.txt @@ -0,0 +1,11 @@ +HTML.Attr.Name.UseCDATA +TYPE: bool +DEFAULT: false +VERSION: 4.0.0 +--DESCRIPTION-- +The W3C specification DTD defines the name attribute to be CDATA, not ID, due +to limitations of DTD. In certain documents, this relaxed behavior is desired, +whether it is to specify duplicate names, or to specify names that would be +illegal IDs (for example, names that begin with a digit.) Set this configuration +directive to true to use the relaxed parsing rules. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.BlockWrapper.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.BlockWrapper.txt new file mode 100644 index 000000000..45ae469ec --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.BlockWrapper.txt @@ -0,0 +1,18 @@ +HTML.BlockWrapper +TYPE: string +VERSION: 1.3.0 +DEFAULT: 'p' +--DESCRIPTION-- + +<p> + String name of element to wrap inline elements that are inside a block + context. This only occurs in the children of blockquote in strict mode. +</p> +<p> + Example: by default value, + <code><blockquote>Foo</blockquote></code> would become + <code><blockquote><p>Foo</p></blockquote></code>. + The <code><p></code> tags can be replaced with whatever you desire, + as long as it is a block level element. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.CoreModules.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.CoreModules.txt new file mode 100644 index 000000000..524618879 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.CoreModules.txt @@ -0,0 +1,23 @@ +HTML.CoreModules +TYPE: lookup +VERSION: 2.0.0 +--DEFAULT-- +array ( + 'Structure' => true, + 'Text' => true, + 'Hypertext' => true, + 'List' => true, + 'NonXMLCommonAttributes' => true, + 'XMLCommonAttributes' => true, + 'CommonAttributes' => true, +) +--DESCRIPTION-- + +<p> + Certain modularized doctypes (XHTML, namely), have certain modules + that must be included for the doctype to be an conforming document + type: put those modules here. By default, XHTML's core modules + are used. You can set this to a blank array to disable core module + protection, but this is not recommended. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.CustomDoctype.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.CustomDoctype.txt new file mode 100644 index 000000000..a64e3d7c3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.CustomDoctype.txt @@ -0,0 +1,9 @@ +HTML.CustomDoctype +TYPE: string/null +VERSION: 2.0.1 +DEFAULT: NULL +--DESCRIPTION-- + +A custom doctype for power-users who defined there own document +type. This directive only applies when %HTML.Doctype is blank. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.DefinitionID.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.DefinitionID.txt new file mode 100644 index 000000000..103db754a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.DefinitionID.txt @@ -0,0 +1,33 @@ +HTML.DefinitionID +TYPE: string/null +DEFAULT: NULL +VERSION: 2.0.0 +--DESCRIPTION-- + +<p> + Unique identifier for a custom-built HTML definition. If you edit + the raw version of the HTMLDefinition, introducing changes that the + configuration object does not reflect, you must specify this variable. + If you change your custom edits, you should change this directive, or + clear your cache. Example: +</p> +<pre> +$config = HTMLPurifier_Config::createDefault(); +$config->set('HTML', 'DefinitionID', '1'); +$def = $config->getHTMLDefinition(); +$def->addAttribute('a', 'tabindex', 'Number'); +</pre> +<p> + In the above example, the configuration is still at the defaults, but + using the advanced API, an extra attribute has been added. The + configuration object normally has no way of knowing that this change + has taken place, so it needs an extra directive: %HTML.DefinitionID. + If someone else attempts to use the default configuration, these two + pieces of code will not clobber each other in the cache, since one has + an extra directive attached to it. +</p> +<p> + You <em>must</em> specify a value to this directive to use the + advanced API features. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.DefinitionRev.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.DefinitionRev.txt new file mode 100644 index 000000000..229ae0267 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.DefinitionRev.txt @@ -0,0 +1,16 @@ +HTML.DefinitionRev +TYPE: int +VERSION: 2.0.0 +DEFAULT: 1 +--DESCRIPTION-- + +<p> + Revision identifier for your custom definition specified in + %HTML.DefinitionID. This serves the same purpose: uniquely identifying + your custom definition, but this one does so in a chronological + context: revision 3 is more up-to-date then revision 2. Thus, when + this gets incremented, the cache handling is smart enough to clean + up any older revisions of your definition as well as flush the + cache. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Doctype.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Doctype.txt new file mode 100644 index 000000000..9dab497f2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Doctype.txt @@ -0,0 +1,11 @@ +HTML.Doctype +TYPE: string/null +DEFAULT: NULL +--DESCRIPTION-- +Doctype to use during filtering. Technically speaking this is not actually +a doctype (as it does not identify a corresponding DTD), but we are using +this name for sake of simplicity. When non-blank, this will override any +older directives like %HTML.XHTML or %HTML.Strict. +--ALLOWED-- +'HTML 4.01 Transitional', 'HTML 4.01 Strict', 'XHTML 1.0 Transitional', 'XHTML 1.0 Strict', 'XHTML 1.1' +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.FlashAllowFullScreen.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.FlashAllowFullScreen.txt new file mode 100644 index 000000000..7878dc0bf --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.FlashAllowFullScreen.txt @@ -0,0 +1,11 @@ +HTML.FlashAllowFullScreen +TYPE: bool +VERSION: 4.2.0 +DEFAULT: false +--DESCRIPTION-- +<p> + Whether or not to permit embedded Flash content from + %HTML.SafeObject to expand to the full screen. Corresponds to + the <code>allowFullScreen</code> parameter. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.ForbiddenAttributes.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.ForbiddenAttributes.txt new file mode 100644 index 000000000..57358f9ba --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.ForbiddenAttributes.txt @@ -0,0 +1,21 @@ +HTML.ForbiddenAttributes +TYPE: lookup +VERSION: 3.1.0 +DEFAULT: array() +--DESCRIPTION-- +<p> + While this directive is similar to %HTML.AllowedAttributes, for + forwards-compatibility with XML, this attribute has a different syntax. Instead of + <code>tag.attr</code>, use <code>tag@attr</code>. To disallow <code>href</code> + attributes in <code>a</code> tags, set this directive to + <code>a@href</code>. You can also disallow an attribute globally with + <code>attr</code> or <code>*@attr</code> (either syntax is fine; the latter + is provided for consistency with %HTML.AllowedAttributes). +</p> +<p> + <strong>Warning:</strong> This directive complements %HTML.ForbiddenElements, + accordingly, check + out that directive for a discussion of why you + should think twice before using this directive. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.ForbiddenElements.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.ForbiddenElements.txt new file mode 100644 index 000000000..93a53e14f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.ForbiddenElements.txt @@ -0,0 +1,20 @@ +HTML.ForbiddenElements +TYPE: lookup +VERSION: 3.1.0 +DEFAULT: array() +--DESCRIPTION-- +<p> + This was, perhaps, the most requested feature ever in HTML + Purifier. Please don't abuse it! This is the logical inverse of + %HTML.AllowedElements, and it will override that directive, or any + other directive. +</p> +<p> + If possible, %HTML.Allowed is recommended over this directive, because it + can sometimes be difficult to tell whether or not you've forbidden all of + the behavior you would like to disallow. If you forbid <code>img</code> + with the expectation of preventing images on your site, you'll be in for + a nasty surprise when people start using the <code>background-image</code> + CSS property. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.MaxImgLength.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.MaxImgLength.txt new file mode 100644 index 000000000..e424c386e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.MaxImgLength.txt @@ -0,0 +1,14 @@ +HTML.MaxImgLength +TYPE: int/null +DEFAULT: 1200 +VERSION: 3.1.1 +--DESCRIPTION-- +<p> + This directive controls the maximum number of pixels in the width and + height attributes in <code>img</code> tags. This is + in place to prevent imagecrash attacks, disable with null at your own risk. + This directive is similar to %CSS.MaxImgLength, and both should be + concurrently edited, although there are + subtle differences in the input format (the HTML max is an integer). +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Nofollow.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Nofollow.txt new file mode 100644 index 000000000..700b30924 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Nofollow.txt @@ -0,0 +1,7 @@ +HTML.Nofollow +TYPE: bool +VERSION: 4.3.0 +DEFAULT: FALSE +--DESCRIPTION-- +If enabled, nofollow rel attributes are added to all outgoing links. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Parent.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Parent.txt new file mode 100644 index 000000000..62e8e160c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Parent.txt @@ -0,0 +1,12 @@ +HTML.Parent +TYPE: string +VERSION: 1.3.0 +DEFAULT: 'div' +--DESCRIPTION-- + +<p> + String name of element that HTML fragment passed to library will be + inserted in. An interesting variation would be using span as the + parent element, meaning that only inline tags would be allowed. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Proprietary.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Proprietary.txt new file mode 100644 index 000000000..dfb720496 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Proprietary.txt @@ -0,0 +1,12 @@ +HTML.Proprietary +TYPE: bool +VERSION: 3.1.0 +DEFAULT: false +--DESCRIPTION-- +<p> + Whether or not to allow proprietary elements and attributes in your + documents, as per <code>HTMLPurifier_HTMLModule_Proprietary</code>. + <strong>Warning:</strong> This can cause your documents to stop + validating! +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeEmbed.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeEmbed.txt new file mode 100644 index 000000000..cdda09a4c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeEmbed.txt @@ -0,0 +1,13 @@ +HTML.SafeEmbed +TYPE: bool +VERSION: 3.1.1 +DEFAULT: false +--DESCRIPTION-- +<p> + Whether or not to permit embed tags in documents, with a number of extra + security features added to prevent script execution. This is similar to + what websites like MySpace do to embed tags. Embed is a proprietary + element and will cause your website to stop validating; you should + see if you can use %Output.FlashCompat with %HTML.SafeObject instead + first.</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeIframe.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeIframe.txt new file mode 100644 index 000000000..5eb6ec2b5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeIframe.txt @@ -0,0 +1,13 @@ +HTML.SafeIframe +TYPE: bool +VERSION: 4.4.0 +DEFAULT: false +--DESCRIPTION-- +<p> + Whether or not to permit iframe tags in untrusted documents. This + directive must be accompanied by a whitelist of permitted iframes, + such as %URI.SafeIframeRegexp, otherwise it will fatally error. + This directive has no effect on strict doctypes, as iframes are not + valid. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeObject.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeObject.txt new file mode 100644 index 000000000..ceb342e22 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.SafeObject.txt @@ -0,0 +1,13 @@ +HTML.SafeObject +TYPE: bool +VERSION: 3.1.1 +DEFAULT: false +--DESCRIPTION-- +<p> + Whether or not to permit object tags in documents, with a number of extra + security features added to prevent script execution. This is similar to + what websites like MySpace do to object tags. You should also enable + %Output.FlashCompat in order to generate Internet Explorer + compatibility code for your object tags. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Strict.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Strict.txt new file mode 100644 index 000000000..a8b1de56b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Strict.txt @@ -0,0 +1,9 @@ +HTML.Strict +TYPE: bool +VERSION: 1.3.0 +DEFAULT: false +DEPRECATED-VERSION: 1.7.0 +DEPRECATED-USE: HTML.Doctype +--DESCRIPTION-- +Determines whether or not to use Transitional (loose) or Strict rulesets. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TargetBlank.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TargetBlank.txt new file mode 100644 index 000000000..587a16778 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TargetBlank.txt @@ -0,0 +1,8 @@ +HTML.TargetBlank +TYPE: bool +VERSION: 4.4.0 +DEFAULT: FALSE +--DESCRIPTION-- +If enabled, <code>target=blank</code> attributes are added to all outgoing links. +(This includes links from an HTTPS version of a page to an HTTP version.) +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyAdd.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyAdd.txt new file mode 100644 index 000000000..b4c271b7f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyAdd.txt @@ -0,0 +1,8 @@ +HTML.TidyAdd +TYPE: lookup +VERSION: 2.0.0 +DEFAULT: array() +--DESCRIPTION-- + +Fixes to add to the default set of Tidy fixes as per your level. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyLevel.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyLevel.txt new file mode 100644 index 000000000..4186ccd0d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyLevel.txt @@ -0,0 +1,24 @@ +HTML.TidyLevel +TYPE: string +VERSION: 2.0.0 +DEFAULT: 'medium' +--DESCRIPTION-- + +<p>General level of cleanliness the Tidy module should enforce. +There are four allowed values:</p> +<dl> + <dt>none</dt> + <dd>No extra tidying should be done</dd> + <dt>light</dt> + <dd>Only fix elements that would be discarded otherwise due to + lack of support in doctype</dd> + <dt>medium</dt> + <dd>Enforce best practices</dd> + <dt>heavy</dt> + <dd>Transform all deprecated elements and attributes to standards + compliant equivalents</dd> +</dl> + +--ALLOWED-- +'none', 'light', 'medium', 'heavy' +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyRemove.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyRemove.txt new file mode 100644 index 000000000..996762bd1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.TidyRemove.txt @@ -0,0 +1,8 @@ +HTML.TidyRemove +TYPE: lookup +VERSION: 2.0.0 +DEFAULT: array() +--DESCRIPTION-- + +Fixes to remove from the default set of Tidy fixes as per your level. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Trusted.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Trusted.txt new file mode 100644 index 000000000..1db9237e9 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.Trusted.txt @@ -0,0 +1,9 @@ +HTML.Trusted +TYPE: bool +VERSION: 2.0.0 +DEFAULT: false +--DESCRIPTION-- +Indicates whether or not the user input is trusted or not. If the input is +trusted, a more expansive set of allowed tags and attributes will be used. +See also %CSS.Trusted. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.XHTML.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.XHTML.txt new file mode 100644 index 000000000..2a47e384f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/HTML.XHTML.txt @@ -0,0 +1,11 @@ +HTML.XHTML +TYPE: bool +DEFAULT: true +VERSION: 1.1.0 +DEPRECATED-VERSION: 1.7.0 +DEPRECATED-USE: HTML.Doctype +--DESCRIPTION-- +Determines whether or not output is XHTML 1.0 or HTML 4.01 flavor. +--ALIASES-- +Core.XHTML +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.CommentScriptContents.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.CommentScriptContents.txt new file mode 100644 index 000000000..08921fde7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.CommentScriptContents.txt @@ -0,0 +1,10 @@ +Output.CommentScriptContents +TYPE: bool +VERSION: 2.0.0 +DEFAULT: true +--DESCRIPTION-- +Determines whether or not HTML Purifier should attempt to fix up the +contents of script tags for legacy browsers with comments. +--ALIASES-- +Core.CommentScriptContents +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.FixInnerHTML.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.FixInnerHTML.txt new file mode 100644 index 000000000..d6f0d9f29 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.FixInnerHTML.txt @@ -0,0 +1,15 @@ +Output.FixInnerHTML +TYPE: bool +VERSION: 4.3.0 +DEFAULT: true +--DESCRIPTION-- +<p> + If true, HTML Purifier will protect against Internet Explorer's + mishandling of the <code>innerHTML</code> attribute by appending + a space to any attribute that does not contain angled brackets, spaces + or quotes, but contains a backtick. This slightly changes the + semantics of any given attribute, so if this is unacceptable and + you do not use <code>innerHTML</code> on any of your pages, you can + turn this directive off. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.FlashCompat.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.FlashCompat.txt new file mode 100644 index 000000000..93398e859 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.FlashCompat.txt @@ -0,0 +1,11 @@ +Output.FlashCompat +TYPE: bool +VERSION: 4.1.0 +DEFAULT: false +--DESCRIPTION-- +<p> + If true, HTML Purifier will generate Internet Explorer compatibility + code for all object code. This is highly recommended if you enable + %HTML.SafeObject. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.Newline.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.Newline.txt new file mode 100644 index 000000000..79f8ad82c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.Newline.txt @@ -0,0 +1,13 @@ +Output.Newline +TYPE: string/null +VERSION: 2.0.1 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + Newline string to format final output with. If left null, HTML Purifier + will auto-detect the default newline type of the system and use that; + you can manually override it here. Remember, \r\n is Windows, \r + is Mac, and \n is Unix. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.SortAttr.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.SortAttr.txt new file mode 100644 index 000000000..232b02362 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.SortAttr.txt @@ -0,0 +1,14 @@ +Output.SortAttr +TYPE: bool +VERSION: 3.2.0 +DEFAULT: false +--DESCRIPTION-- +<p> + If true, HTML Purifier will sort attributes by name before writing them back + to the document, converting a tag like: <code><el b="" a="" c="" /></code> + to <code><el a="" b="" c="" /></code>. This is a workaround for + a bug in FCKeditor which causes it to swap attributes order, adding noise + to text diffs. If you're not seeing this bug, chances are, you don't need + this directive. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.TidyFormat.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.TidyFormat.txt new file mode 100644 index 000000000..06bab00a0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Output.TidyFormat.txt @@ -0,0 +1,25 @@ +Output.TidyFormat +TYPE: bool +VERSION: 1.1.1 +DEFAULT: false +--DESCRIPTION-- +<p> + Determines whether or not to run Tidy on the final output for pretty + formatting reasons, such as indentation and wrap. +</p> +<p> + This can greatly improve readability for editors who are hand-editing + the HTML, but is by no means necessary as HTML Purifier has already + fixed all major errors the HTML may have had. Tidy is a non-default + extension, and this directive will silently fail if Tidy is not + available. +</p> +<p> + If you are looking to make the overall look of your page's source + better, I recommend running Tidy on the entire page rather than just + user-content (after all, the indentation relative to the containing + blocks will be incorrect). +</p> +--ALIASES-- +Core.TidyFormat +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Test.ForceNoIconv.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Test.ForceNoIconv.txt new file mode 100644 index 000000000..071bc0295 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/Test.ForceNoIconv.txt @@ -0,0 +1,7 @@ +Test.ForceNoIconv +TYPE: bool +DEFAULT: false +--DESCRIPTION-- +When set to true, HTMLPurifier_Encoder will act as if iconv does not exist +and use only pure PHP implementations. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt new file mode 100644 index 000000000..666635a5f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.AllowedSchemes.txt @@ -0,0 +1,17 @@ +URI.AllowedSchemes +TYPE: lookup +--DEFAULT-- +array ( + 'http' => true, + 'https' => true, + 'mailto' => true, + 'ftp' => true, + 'nntp' => true, + 'news' => true, +) +--DESCRIPTION-- +Whitelist that defines the schemes that a URI is allowed to have. This +prevents XSS attacks from using pseudo-schemes like javascript or mocha. +There is also support for the <code>data</code> and <code>file</code> +URI schemes, but they are not enabled by default. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Base.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Base.txt new file mode 100644 index 000000000..876f0680c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Base.txt @@ -0,0 +1,17 @@ +URI.Base +TYPE: string/null +VERSION: 2.1.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + The base URI is the URI of the document this purified HTML will be + inserted into. This information is important if HTML Purifier needs + to calculate absolute URIs from relative URIs, such as when %URI.MakeAbsolute + is on. You may use a non-absolute URI for this value, but behavior + may vary (%URI.MakeAbsolute deals nicely with both absolute and + relative paths, but forwards-compatibility is not guaranteed). + <strong>Warning:</strong> If set, the scheme on this URI + overrides the one specified by %URI.DefaultScheme. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt new file mode 100644 index 000000000..728e378cb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefaultScheme.txt @@ -0,0 +1,10 @@ +URI.DefaultScheme +TYPE: string +DEFAULT: 'http' +--DESCRIPTION-- + +<p> + Defines through what scheme the output will be served, in order to + select the proper object validator when no scheme information is present. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefinitionID.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefinitionID.txt new file mode 100644 index 000000000..f05312ba8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefinitionID.txt @@ -0,0 +1,11 @@ +URI.DefinitionID +TYPE: string/null +VERSION: 2.1.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + Unique identifier for a custom-built URI definition. If you want + to add custom URIFilters, you must specify this value. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefinitionRev.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefinitionRev.txt new file mode 100644 index 000000000..80cfea93f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DefinitionRev.txt @@ -0,0 +1,11 @@ +URI.DefinitionRev +TYPE: int +VERSION: 2.1.0 +DEFAULT: 1 +--DESCRIPTION-- + +<p> + Revision identifier for your custom definition. See + %HTML.DefinitionRev for details. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Disable.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Disable.txt new file mode 100644 index 000000000..71ce025a2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Disable.txt @@ -0,0 +1,14 @@ +URI.Disable +TYPE: bool +VERSION: 1.3.0 +DEFAULT: false +--DESCRIPTION-- + +<p> + Disables all URIs in all forms. Not sure why you'd want to do that + (after all, the Internet's founded on the notion of a hyperlink). +</p> + +--ALIASES-- +Attr.DisableURI +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableExternal.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableExternal.txt new file mode 100644 index 000000000..13c122c8c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableExternal.txt @@ -0,0 +1,11 @@ +URI.DisableExternal +TYPE: bool +VERSION: 1.2.0 +DEFAULT: false +--DESCRIPTION-- +Disables links to external websites. This is a highly effective anti-spam +and anti-pagerank-leech measure, but comes at a hefty price: nolinks or +images outside of your domain will be allowed. Non-linkified URIs will +still be preserved. If you want to be able to link to subdomains or use +absolute URIs, specify %URI.Host for your website. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableExternalResources.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableExternalResources.txt new file mode 100644 index 000000000..abcc1efd6 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableExternalResources.txt @@ -0,0 +1,13 @@ +URI.DisableExternalResources +TYPE: bool +VERSION: 1.3.0 +DEFAULT: false +--DESCRIPTION-- +Disables the embedding of external resources, preventing users from +embedding things like images from other hosts. This prevents access +tracking (good for email viewers), bandwidth leeching, cross-site request +forging, goatse.cx posting, and other nasties, but also results in a loss +of end-user functionality (they can't directly post a pic they posted from +Flickr anymore). Use it if you don't have a robust user-content moderation +team. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableResources.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableResources.txt new file mode 100644 index 000000000..f891de499 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.DisableResources.txt @@ -0,0 +1,15 @@ +URI.DisableResources +TYPE: bool +VERSION: 4.2.0 +DEFAULT: false +--DESCRIPTION-- +<p> + Disables embedding resources, essentially meaning no pictures. You can + still link to them though. See %URI.DisableExternalResources for why + this might be a good idea. +</p> +<p> + <em>Note:</em> While this directive has been available since 1.3.0, + it didn't actually start doing anything until 4.2.0. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Host.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Host.txt new file mode 100644 index 000000000..ee83b121d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Host.txt @@ -0,0 +1,19 @@ +URI.Host +TYPE: string/null +VERSION: 1.2.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + Defines the domain name of the server, so we can determine whether or + an absolute URI is from your website or not. Not strictly necessary, + as users should be using relative URIs to reference resources on your + website. It will, however, let you use absolute URIs to link to + subdomains of the domain you post here: i.e. example.com will allow + sub.example.com. However, higher up domains will still be excluded: + if you set %URI.Host to sub.example.com, example.com will be blocked. + <strong>Note:</strong> This directive overrides %URI.Base because + a given page may be on a sub-domain, but you wish HTML Purifier to be + more relaxed and allow some of the parent domains too. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.HostBlacklist.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.HostBlacklist.txt new file mode 100644 index 000000000..0b6df7625 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.HostBlacklist.txt @@ -0,0 +1,9 @@ +URI.HostBlacklist +TYPE: list +VERSION: 1.3.0 +DEFAULT: array() +--DESCRIPTION-- +List of strings that are forbidden in the host of any URI. Use it to kill +domain names of spam, etc. Note that it will catch anything in the domain, +so <tt>moo.com</tt> will catch <tt>moo.com.example.com</tt>. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MakeAbsolute.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MakeAbsolute.txt new file mode 100644 index 000000000..4214900a5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MakeAbsolute.txt @@ -0,0 +1,13 @@ +URI.MakeAbsolute +TYPE: bool +VERSION: 2.1.0 +DEFAULT: false +--DESCRIPTION-- + +<p> + Converts all URIs into absolute forms. This is useful when the HTML + being filtered assumes a specific base path, but will actually be + viewed in a different context (and setting an alternate base URI is + not possible). %URI.Base must be set for this directive to work. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt new file mode 100644 index 000000000..58c81dcc4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.Munge.txt @@ -0,0 +1,83 @@ +URI.Munge +TYPE: string/null +VERSION: 1.3.0 +DEFAULT: NULL +--DESCRIPTION-- + +<p> + Munges all browsable (usually http, https and ftp) + absolute URIs into another URI, usually a URI redirection service. + This directive accepts a URI, formatted with a <code>%s</code> where + the url-encoded original URI should be inserted (sample: + <code>http://www.google.com/url?q=%s</code>). +</p> +<p> + Uses for this directive: +</p> +<ul> + <li> + 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). <strong>Notice</strong>: + Many security experts believe that this form of protection does not deter spam-bots. + </li> + <li> + 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. + </li> +</ul> +<p> + 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 <code>meta</code> tags. To revert to + previous behavior, please use %URI.MungeResources. +</p> +<p> + 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. +</p> +<p> + Starting with HTML Purifier 3.1.1, there is also these substitutions: +</p> +<table> + <thead> + <tr> + <th>Key</th> + <th>Description</th> + <th>Example <code><a href=""></code></th> + </tr> + </thead> + <tbody> + <tr> + <td>%r</td> + <td>1 - The URI embeds a resource<br />(blank) - The URI is merely a link</td> + <td></td> + </tr> + <tr> + <td>%n</td> + <td>The name of the tag this URI came from</td> + <td>a</td> + </tr> + <tr> + <td>%m</td> + <td>The name of the attribute this URI came from</td> + <td>href</td> + </tr> + <tr> + <td>%p</td> + <td>The name of the CSS property this URI came from, or blank if irrelevant</td> + <td></td> + </tr> + </tbody> +</table> +<p> + 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. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MungeResources.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MungeResources.txt new file mode 100644 index 000000000..6fce0fdc3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MungeResources.txt @@ -0,0 +1,17 @@ +URI.MungeResources +TYPE: bool +VERSION: 3.1.1 +DEFAULT: false +--DESCRIPTION-- +<p> + If true, any URI munging directives like %URI.Munge + will also apply to embedded resources, such as <code><img src=""></code>. + Be careful enabling this directive if you have a redirector script + that does not use the <code>Location</code> HTTP header; all of your images + and other embedded resources will break. +</p> +<p> + <strong>Warning:</strong> It is strongly advised you use this in conjunction + %URI.MungeSecretKey to mitigate the security risk of an open redirector. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MungeSecretKey.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MungeSecretKey.txt new file mode 100644 index 000000000..0d00f62ea --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.MungeSecretKey.txt @@ -0,0 +1,30 @@ +URI.MungeSecretKey +TYPE: string/null +VERSION: 3.1.1 +DEFAULT: NULL +--DESCRIPTION-- +<p> + This directive enables secure checksum generation along with %URI.Munge. + It should be set to a secure key that is not shared with anyone else. + The checksum can be placed in the URI using %t. Use of this checksum + affords an additional level of protection by allowing a redirector + to check if a URI has passed through HTML Purifier with this line: +</p> + +<pre>$checksum === sha1($secret_key . ':' . $url)</pre> + +<p> + If the output is TRUE, the redirector script should accept the URI. +</p> + +<p> + Please note that it would still be possible for an attacker to procure + secure hashes en-mass by abusing your website's Preview feature or the + like, but this service affords an additional level of protection + that should be combined with website blacklisting. +</p> + +<p> + Remember this has no effect if %URI.Munge is not on. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.OverrideAllowedSchemes.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.OverrideAllowedSchemes.txt new file mode 100644 index 000000000..23331a4e7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.OverrideAllowedSchemes.txt @@ -0,0 +1,9 @@ +URI.OverrideAllowedSchemes +TYPE: bool +DEFAULT: true +--DESCRIPTION-- +If this is set to true (which it is by default), you can override +%URI.AllowedSchemes by simply registering a HTMLPurifier_URIScheme to the +registry. If false, you will also have to update that directive in order +to add more schemes. +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.SafeIframeRegexp.txt b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.SafeIframeRegexp.txt new file mode 100644 index 000000000..79084832b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/URI.SafeIframeRegexp.txt @@ -0,0 +1,22 @@ +URI.SafeIframeRegexp +TYPE: string/null +VERSION: 4.4.0 +DEFAULT: NULL +--DESCRIPTION-- +<p> + A PCRE regular expression that will be matched against an iframe URI. This is + a relatively inflexible scheme, but works well enough for the most common + use-case of iframes: embedded video. This directive only has an effect if + %HTML.SafeIframe is enabled. Here are some example values: +</p> +<ul> + <li><code>%^http://www.youtube.com/embed/%</code> - Allow YouTube videos</li> + <li><code>%^http://player.vimeo.com/video/%</code> - Allow Vimeo videos</li> + <li><code>%^http://(www.youtube.com/embed/|player.vimeo.com/video/)%</code> - Allow both</li> +</ul> +<p> + Note that this directive does not give you enough granularity to, say, disable + all <code>autoplay</code> videos. Pipe up on the HTML Purifier forums if this + is a capability you want. +</p> +--# vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/info.ini b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/info.ini new file mode 100644 index 000000000..5de4505e1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ConfigSchema/schema/info.ini @@ -0,0 +1,3 @@ +name = "HTML Purifier" + +; vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ContentSets.php b/lib/htmlpurifier/library/HTMLPurifier/ContentSets.php new file mode 100644 index 000000000..3b6e96f5f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ContentSets.php @@ -0,0 +1,155 @@ +<?php + +/** + * @todo Unit test + */ +class HTMLPurifier_ContentSets +{ + + /** + * List of content set strings (pipe seperators) indexed by name. + */ + public $info = array(); + + /** + * List of content set lookups (element => true) indexed by name. + * @note This is in HTMLPurifier_HTMLDefinition->info_content_sets + */ + public $lookup = array(); + + /** + * Synchronized list of defined content sets (keys of info) + */ + protected $keys = array(); + /** + * Synchronized list of defined content values (values of info) + */ + protected $values = array(); + + /** + * Merges in module's content sets, expands identifiers in the content + * sets and populates the keys, values and lookup member variables. + * @param $modules List of HTMLPurifier_HTMLModule + */ + public function __construct($modules) { + if (!is_array($modules)) $modules = array($modules); + // populate content_sets based on module hints + // sorry, no way of overloading + foreach ($modules as $module_i => $module) { + foreach ($module->content_sets as $key => $value) { + $temp = $this->convertToLookup($value); + if (isset($this->lookup[$key])) { + // add it into the existing content set + $this->lookup[$key] = array_merge($this->lookup[$key], $temp); + } else { + $this->lookup[$key] = $temp; + } + } + } + $old_lookup = false; + while ($old_lookup !== $this->lookup) { + $old_lookup = $this->lookup; + foreach ($this->lookup as $i => $set) { + $add = array(); + foreach ($set as $element => $x) { + if (isset($this->lookup[$element])) { + $add += $this->lookup[$element]; + unset($this->lookup[$i][$element]); + } + } + $this->lookup[$i] += $add; + } + } + + foreach ($this->lookup as $key => $lookup) { + $this->info[$key] = implode(' | ', array_keys($lookup)); + } + $this->keys = array_keys($this->info); + $this->values = array_values($this->info); + } + + /** + * Accepts a definition; generates and assigns a ChildDef for it + * @param $def HTMLPurifier_ElementDef reference + * @param $module Module that defined the ElementDef + */ + public function generateChildDef(&$def, $module) { + if (!empty($def->child)) return; // already done! + $content_model = $def->content_model; + if (is_string($content_model)) { + // Assume that $this->keys is alphanumeric + $def->content_model = preg_replace_callback( + '/\b(' . implode('|', $this->keys) . ')\b/', + array($this, 'generateChildDefCallback'), + $content_model + ); + //$def->content_model = str_replace( + // $this->keys, $this->values, $content_model); + } + $def->child = $this->getChildDef($def, $module); + } + + public function generateChildDefCallback($matches) { + return $this->info[$matches[0]]; + } + + /** + * Instantiates a ChildDef based on content_model and content_model_type + * member variables in HTMLPurifier_ElementDef + * @note This will also defer to modules for custom HTMLPurifier_ChildDef + * subclasses that need content set expansion + * @param $def HTMLPurifier_ElementDef to have ChildDef extracted + * @return HTMLPurifier_ChildDef corresponding to ElementDef + */ + public function getChildDef($def, $module) { + $value = $def->content_model; + if (is_object($value)) { + trigger_error( + 'Literal object child definitions should be stored in '. + 'ElementDef->child not ElementDef->content_model', + E_USER_NOTICE + ); + return $value; + } + switch ($def->content_model_type) { + case 'required': + return new HTMLPurifier_ChildDef_Required($value); + case 'optional': + return new HTMLPurifier_ChildDef_Optional($value); + case 'empty': + return new HTMLPurifier_ChildDef_Empty(); + case 'custom': + return new HTMLPurifier_ChildDef_Custom($value); + } + // defer to its module + $return = false; + if ($module->defines_child_def) { // save a func call + $return = $module->getChildDef($def); + } + if ($return !== false) return $return; + // error-out + trigger_error( + 'Could not determine which ChildDef class to instantiate', + E_USER_ERROR + ); + return false; + } + + /** + * Converts a string list of elements separated by pipes into + * a lookup array. + * @param $string List of elements + * @return Lookup array of elements + */ + protected function convertToLookup($string) { + $array = explode('|', str_replace(' ', '', $string)); + $ret = array(); + foreach ($array as $i => $k) { + $ret[$k] = true; + } + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Context.php b/lib/htmlpurifier/library/HTMLPurifier/Context.php new file mode 100644 index 000000000..9ddf0c547 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Context.php @@ -0,0 +1,82 @@ +<?php + +/** + * Registry object that contains information about the current context. + * @warning Is a bit buggy when variables are set to null: it thinks + * they don't exist! So use false instead, please. + * @note Since the variables Context deals with may not be objects, + * references are very important here! Do not remove! + */ +class HTMLPurifier_Context +{ + + /** + * Private array that stores the references. + */ + private $_storage = array(); + + /** + * Registers a variable into the context. + * @param $name String name + * @param $ref Reference to variable to be registered + */ + public function register($name, &$ref) { + if (isset($this->_storage[$name])) { + trigger_error("Name $name produces collision, cannot re-register", + E_USER_ERROR); + return; + } + $this->_storage[$name] =& $ref; + } + + /** + * Retrieves a variable reference from the context. + * @param $name String name + * @param $ignore_error Boolean whether or not to ignore error + */ + public function &get($name, $ignore_error = false) { + if (!isset($this->_storage[$name])) { + if (!$ignore_error) { + trigger_error("Attempted to retrieve non-existent variable $name", + E_USER_ERROR); + } + $var = null; // so we can return by reference + return $var; + } + return $this->_storage[$name]; + } + + /** + * Destorys a variable in the context. + * @param $name String name + */ + public function destroy($name) { + if (!isset($this->_storage[$name])) { + trigger_error("Attempted to destroy non-existent variable $name", + E_USER_ERROR); + return; + } + unset($this->_storage[$name]); + } + + /** + * Checks whether or not the variable exists. + * @param $name String name + */ + public function exists($name) { + return isset($this->_storage[$name]); + } + + /** + * Loads a series of variables from an associative array + * @param $context_array Assoc array of variables to load + */ + public function loadArray($context_array) { + foreach ($context_array as $key => $discard) { + $this->register($key, $context_array[$key]); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Definition.php b/lib/htmlpurifier/library/HTMLPurifier/Definition.php new file mode 100644 index 000000000..c7f82eba4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Definition.php @@ -0,0 +1,50 @@ +<?php + +/** + * Super-class for definition datatype objects, implements serialization + * functions for the class. + */ +abstract class HTMLPurifier_Definition +{ + + /** + * Has setup() been called yet? + */ + public $setup = false; + + /** + * If true, write out the final definition object to the cache after + * setup. This will be true only if all invocations to get a raw + * definition object are also optimized. This does not cause file + * system thrashing because on subsequent calls the cached object + * is used and any writes to the raw definition object are short + * circuited. See enduser-customize.html for the high-level + * picture. + */ + public $optimized = null; + + /** + * What type of definition is it? + */ + public $type; + + /** + * Sets up the definition object into the final form, something + * not done by the constructor + * @param $config HTMLPurifier_Config instance + */ + abstract protected function doSetup($config); + + /** + * Setup function that aborts if already setup + * @param $config HTMLPurifier_Config instance + */ + public function setup($config) { + if ($this->setup) return; + $this->setup = true; + $this->doSetup($config); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache.php new file mode 100644 index 000000000..c6e1e388c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache.php @@ -0,0 +1,108 @@ +<?php + +/** + * Abstract class representing Definition cache managers that implements + * useful common methods and is a factory. + * @todo Create a separate maintenance file advanced users can use to + * cache their custom HTMLDefinition, which can be loaded + * via a configuration directive + * @todo Implement memcached + */ +abstract class HTMLPurifier_DefinitionCache +{ + + public $type; + + /** + * @param $name Type of definition objects this instance of the + * cache will handle. + */ + public function __construct($type) { + $this->type = $type; + } + + /** + * Generates a unique identifier for a particular configuration + * @param Instance of HTMLPurifier_Config + */ + public function generateKey($config) { + return $config->version . ',' . // possibly replace with function calls + $config->getBatchSerial($this->type) . ',' . + $config->get($this->type . '.DefinitionRev'); + } + + /** + * Tests whether or not a key is old with respect to the configuration's + * version and revision number. + * @param $key Key to test + * @param $config Instance of HTMLPurifier_Config to test against + */ + public function isOld($key, $config) { + if (substr_count($key, ',') < 2) return true; + list($version, $hash, $revision) = explode(',', $key, 3); + $compare = version_compare($version, $config->version); + // version mismatch, is always old + if ($compare != 0) return true; + // versions match, ids match, check revision number + if ( + $hash == $config->getBatchSerial($this->type) && + $revision < $config->get($this->type . '.DefinitionRev') + ) return true; + return false; + } + + /** + * Checks if a definition's type jives with the cache's type + * @note Throws an error on failure + * @param $def Definition object to check + * @return Boolean true if good, false if not + */ + public function checkDefType($def) { + if ($def->type !== $this->type) { + trigger_error("Cannot use definition of type {$def->type} in cache for {$this->type}"); + return false; + } + return true; + } + + /** + * Adds a definition object to the cache + */ + abstract public function add($def, $config); + + /** + * Unconditionally saves a definition object to the cache + */ + abstract public function set($def, $config); + + /** + * Replace an object in the cache + */ + abstract public function replace($def, $config); + + /** + * Retrieves a definition object from the cache + */ + abstract public function get($config); + + /** + * Removes a definition object to the cache + */ + abstract public function remove($config); + + /** + * Clears all objects from cache + */ + abstract public function flush($config); + + /** + * Clears all expired (older version or revision) objects from cache + * @note Be carefuly implementing this method as flush. Flush must + * not interfere with other Definition types, and cleanup() + * should not be repeatedly called by userland code. + */ + abstract public function cleanup($config); + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator.php new file mode 100644 index 000000000..b0fb6d0cd --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator.php @@ -0,0 +1,62 @@ +<?php + +class HTMLPurifier_DefinitionCache_Decorator extends HTMLPurifier_DefinitionCache +{ + + /** + * Cache object we are decorating + */ + public $cache; + + public function __construct() {} + + /** + * Lazy decorator function + * @param $cache Reference to cache object to decorate + */ + public function decorate(&$cache) { + $decorator = $this->copy(); + // reference is necessary for mocks in PHP 4 + $decorator->cache =& $cache; + $decorator->type = $cache->type; + return $decorator; + } + + /** + * Cross-compatible clone substitute + */ + public function copy() { + return new HTMLPurifier_DefinitionCache_Decorator(); + } + + public function add($def, $config) { + return $this->cache->add($def, $config); + } + + public function set($def, $config) { + return $this->cache->set($def, $config); + } + + public function replace($def, $config) { + return $this->cache->replace($def, $config); + } + + public function get($config) { + return $this->cache->get($config); + } + + public function remove($config) { + return $this->cache->remove($config); + } + + public function flush($config) { + return $this->cache->flush($config); + } + + public function cleanup($config) { + return $this->cache->cleanup($config); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Cleanup.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Cleanup.php new file mode 100644 index 000000000..d4cc35c4b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Cleanup.php @@ -0,0 +1,43 @@ +<?php + +/** + * Definition cache decorator class that cleans up the cache + * whenever there is a cache miss. + */ +class HTMLPurifier_DefinitionCache_Decorator_Cleanup extends + HTMLPurifier_DefinitionCache_Decorator +{ + + public $name = 'Cleanup'; + + public function copy() { + return new HTMLPurifier_DefinitionCache_Decorator_Cleanup(); + } + + public function add($def, $config) { + $status = parent::add($def, $config); + if (!$status) parent::cleanup($config); + return $status; + } + + public function set($def, $config) { + $status = parent::set($def, $config); + if (!$status) parent::cleanup($config); + return $status; + } + + public function replace($def, $config) { + $status = parent::replace($def, $config); + if (!$status) parent::cleanup($config); + return $status; + } + + public function get($config) { + $ret = parent::get($config); + if (!$ret) parent::cleanup($config); + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Memory.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Memory.php new file mode 100644 index 000000000..18f16d32b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Memory.php @@ -0,0 +1,46 @@ +<?php + +/** + * Definition cache decorator class that saves all cache retrievals + * to PHP's memory; good for unit tests or circumstances where + * there are lots of configuration objects floating around. + */ +class HTMLPurifier_DefinitionCache_Decorator_Memory extends + HTMLPurifier_DefinitionCache_Decorator +{ + + protected $definitions; + public $name = 'Memory'; + + public function copy() { + return new HTMLPurifier_DefinitionCache_Decorator_Memory(); + } + + public function add($def, $config) { + $status = parent::add($def, $config); + if ($status) $this->definitions[$this->generateKey($config)] = $def; + return $status; + } + + public function set($def, $config) { + $status = parent::set($def, $config); + if ($status) $this->definitions[$this->generateKey($config)] = $def; + return $status; + } + + public function replace($def, $config) { + $status = parent::replace($def, $config); + if ($status) $this->definitions[$this->generateKey($config)] = $def; + return $status; + } + + public function get($config) { + $key = $this->generateKey($config); + if (isset($this->definitions[$key])) return $this->definitions[$key]; + $this->definitions[$key] = parent::get($config); + return $this->definitions[$key]; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Template.php.in b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Template.php.in new file mode 100644 index 000000000..21a8fcfda --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Decorator/Template.php.in @@ -0,0 +1,47 @@ +<?php + +require_once 'HTMLPurifier/DefinitionCache/Decorator.php'; + +/** + * Definition cache decorator template. + */ +class HTMLPurifier_DefinitionCache_Decorator_Template extends + HTMLPurifier_DefinitionCache_Decorator +{ + + var $name = 'Template'; // replace this + + function copy() { + // replace class name with yours + return new HTMLPurifier_DefinitionCache_Decorator_Template(); + } + + // remove methods you don't need + + function add($def, $config) { + return parent::add($def, $config); + } + + function set($def, $config) { + return parent::set($def, $config); + } + + function replace($def, $config) { + return parent::replace($def, $config); + } + + function get($config) { + return parent::get($config); + } + + function flush() { + return parent::flush(); + } + + function cleanup($config) { + return parent::cleanup($config); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Null.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Null.php new file mode 100644 index 000000000..41d97e734 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Null.php @@ -0,0 +1,39 @@ +<?php + +/** + * Null cache object to use when no caching is on. + */ +class HTMLPurifier_DefinitionCache_Null extends HTMLPurifier_DefinitionCache +{ + + public function add($def, $config) { + return false; + } + + public function set($def, $config) { + return false; + } + + public function replace($def, $config) { + return false; + } + + public function remove($config) { + return false; + } + + public function get($config) { + return false; + } + + public function flush($config) { + return false; + } + + public function cleanup($config) { + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Serializer.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Serializer.php new file mode 100644 index 000000000..73d5e90a6 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Serializer.php @@ -0,0 +1,191 @@ +<?php + +class HTMLPurifier_DefinitionCache_Serializer extends + HTMLPurifier_DefinitionCache +{ + + public function add($def, $config) { + if (!$this->checkDefType($def)) return; + $file = $this->generateFilePath($config); + if (file_exists($file)) return false; + if (!$this->_prepareDir($config)) return false; + return $this->_write($file, serialize($def), $config); + } + + public function set($def, $config) { + if (!$this->checkDefType($def)) return; + $file = $this->generateFilePath($config); + if (!$this->_prepareDir($config)) return false; + return $this->_write($file, serialize($def), $config); + } + + public function replace($def, $config) { + if (!$this->checkDefType($def)) return; + $file = $this->generateFilePath($config); + if (!file_exists($file)) return false; + if (!$this->_prepareDir($config)) return false; + return $this->_write($file, serialize($def), $config); + } + + public function get($config) { + $file = $this->generateFilePath($config); + if (!file_exists($file)) return false; + return unserialize(file_get_contents($file)); + } + + public function remove($config) { + $file = $this->generateFilePath($config); + if (!file_exists($file)) return false; + return unlink($file); + } + + public function flush($config) { + if (!$this->_prepareDir($config)) return false; + $dir = $this->generateDirectoryPath($config); + $dh = opendir($dir); + while (false !== ($filename = readdir($dh))) { + if (empty($filename)) continue; + if ($filename[0] === '.') continue; + unlink($dir . '/' . $filename); + } + } + + public function cleanup($config) { + if (!$this->_prepareDir($config)) return false; + $dir = $this->generateDirectoryPath($config); + $dh = opendir($dir); + while (false !== ($filename = readdir($dh))) { + if (empty($filename)) continue; + if ($filename[0] === '.') continue; + $key = substr($filename, 0, strlen($filename) - 4); + if ($this->isOld($key, $config)) unlink($dir . '/' . $filename); + } + } + + /** + * Generates the file path to the serial file corresponding to + * the configuration and definition name + * @todo Make protected + */ + public function generateFilePath($config) { + $key = $this->generateKey($config); + return $this->generateDirectoryPath($config) . '/' . $key . '.ser'; + } + + /** + * Generates the path to the directory contain this cache's serial files + * @note No trailing slash + * @todo Make protected + */ + public function generateDirectoryPath($config) { + $base = $this->generateBaseDirectoryPath($config); + return $base . '/' . $this->type; + } + + /** + * Generates path to base directory that contains all definition type + * serials + * @todo Make protected + */ + public function generateBaseDirectoryPath($config) { + $base = $config->get('Cache.SerializerPath'); + $base = is_null($base) ? HTMLPURIFIER_PREFIX . '/HTMLPurifier/DefinitionCache/Serializer' : $base; + return $base; + } + + /** + * Convenience wrapper function for file_put_contents + * @param $file File name to write to + * @param $data Data to write into file + * @param $config Config object + * @return Number of bytes written if success, or false if failure. + */ + private function _write($file, $data, $config) { + $result = file_put_contents($file, $data); + if ($result !== false) { + // set permissions of the new file (no execute) + $chmod = $config->get('Cache.SerializerPermissions'); + if (!$chmod) { + $chmod = 0644; // invalid config or simpletest + } + $chmod = $chmod & 0666; + chmod($file, $chmod); + } + return $result; + } + + /** + * Prepares the directory that this type stores the serials in + * @param $config Config object + * @return True if successful + */ + private function _prepareDir($config) { + $directory = $this->generateDirectoryPath($config); + $chmod = $config->get('Cache.SerializerPermissions'); + if (!$chmod) { + $chmod = 0755; // invalid config or simpletest + } + if (!is_dir($directory)) { + $base = $this->generateBaseDirectoryPath($config); + if (!is_dir($base)) { + trigger_error('Base directory '.$base.' does not exist, + please create or change using %Cache.SerializerPath', + E_USER_WARNING); + return false; + } elseif (!$this->_testPermissions($base, $chmod)) { + return false; + } + $old = umask(0000); + mkdir($directory, $chmod); + umask($old); + } elseif (!$this->_testPermissions($directory, $chmod)) { + return false; + } + return true; + } + + /** + * Tests permissions on a directory and throws out friendly + * error messages and attempts to chmod it itself if possible + * @param $dir Directory path + * @param $chmod Permissions + * @return True if directory writable + */ + private function _testPermissions($dir, $chmod) { + // early abort, if it is writable, everything is hunky-dory + if (is_writable($dir)) return true; + if (!is_dir($dir)) { + // generally, you'll want to handle this beforehand + // so a more specific error message can be given + trigger_error('Directory '.$dir.' does not exist', + E_USER_WARNING); + return false; + } + if (function_exists('posix_getuid')) { + // POSIX system, we can give more specific advice + if (fileowner($dir) === posix_getuid()) { + // we can chmod it ourselves + $chmod = $chmod | 0700; + if (chmod($dir, $chmod)) return true; + } elseif (filegroup($dir) === posix_getgid()) { + $chmod = $chmod | 0070; + } else { + // PHP's probably running as nobody, so we'll + // need to give global permissions + $chmod = $chmod | 0777; + } + trigger_error('Directory '.$dir.' not writable, '. + 'please chmod to ' . decoct($chmod), + E_USER_WARNING); + } else { + // generic error message + trigger_error('Directory '.$dir.' not writable, '. + 'please alter file permissions', + E_USER_WARNING); + } + return false; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Serializer/README b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Serializer/README new file mode 100755 index 000000000..2e35c1c3d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCache/Serializer/README @@ -0,0 +1,3 @@ +This is a dummy file to prevent Git from ignoring this empty directory. + + vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DefinitionCacheFactory.php b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCacheFactory.php new file mode 100644 index 000000000..a6ead6281 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DefinitionCacheFactory.php @@ -0,0 +1,91 @@ +<?php + +/** + * Responsible for creating definition caches. + */ +class HTMLPurifier_DefinitionCacheFactory +{ + + protected $caches = array('Serializer' => array()); + protected $implementations = array(); + protected $decorators = array(); + + /** + * Initialize default decorators + */ + public function setup() { + $this->addDecorator('Cleanup'); + } + + /** + * Retrieves an instance of global definition cache factory. + */ + public static function instance($prototype = null) { + static $instance; + if ($prototype !== null) { + $instance = $prototype; + } elseif ($instance === null || $prototype === true) { + $instance = new HTMLPurifier_DefinitionCacheFactory(); + $instance->setup(); + } + return $instance; + } + + /** + * Registers a new definition cache object + * @param $short Short name of cache object, for reference + * @param $long Full class name of cache object, for construction + */ + public function register($short, $long) { + $this->implementations[$short] = $long; + } + + /** + * Factory method that creates a cache object based on configuration + * @param $name Name of definitions handled by cache + * @param $config Instance of HTMLPurifier_Config + */ + public function create($type, $config) { + $method = $config->get('Cache.DefinitionImpl'); + if ($method === null) { + return new HTMLPurifier_DefinitionCache_Null($type); + } + if (!empty($this->caches[$method][$type])) { + return $this->caches[$method][$type]; + } + if ( + isset($this->implementations[$method]) && + class_exists($class = $this->implementations[$method], false) + ) { + $cache = new $class($type); + } else { + if ($method != 'Serializer') { + trigger_error("Unrecognized DefinitionCache $method, using Serializer instead", E_USER_WARNING); + } + $cache = new HTMLPurifier_DefinitionCache_Serializer($type); + } + foreach ($this->decorators as $decorator) { + $new_cache = $decorator->decorate($cache); + // prevent infinite recursion in PHP 4 + unset($cache); + $cache = $new_cache; + } + $this->caches[$method][$type] = $cache; + return $this->caches[$method][$type]; + } + + /** + * Registers a decorator to add to all new cache objects + * @param + */ + public function addDecorator($decorator) { + if (is_string($decorator)) { + $class = "HTMLPurifier_DefinitionCache_Decorator_$decorator"; + $decorator = new $class; + } + $this->decorators[$decorator->name] = $decorator; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Doctype.php b/lib/htmlpurifier/library/HTMLPurifier/Doctype.php new file mode 100644 index 000000000..1e3c574c0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Doctype.php @@ -0,0 +1,60 @@ +<?php + +/** + * Represents a document type, contains information on which modules + * need to be loaded. + * @note This class is inspected by Printer_HTMLDefinition->renderDoctype. + * If structure changes, please update that function. + */ +class HTMLPurifier_Doctype +{ + /** + * Full name of doctype + */ + public $name; + + /** + * List of standard modules (string identifiers or literal objects) + * that this doctype uses + */ + public $modules = array(); + + /** + * List of modules to use for tidying up code + */ + public $tidyModules = array(); + + /** + * Is the language derived from XML (i.e. XHTML)? + */ + public $xml = true; + + /** + * List of aliases for this doctype + */ + public $aliases = array(); + + /** + * Public DTD identifier + */ + public $dtdPublic; + + /** + * System DTD identifier + */ + public $dtdSystem; + + public function __construct($name = null, $xml = true, $modules = array(), + $tidyModules = array(), $aliases = array(), $dtd_public = null, $dtd_system = null + ) { + $this->name = $name; + $this->xml = $xml; + $this->modules = $modules; + $this->tidyModules = $tidyModules; + $this->aliases = $aliases; + $this->dtdPublic = $dtd_public; + $this->dtdSystem = $dtd_system; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/DoctypeRegistry.php b/lib/htmlpurifier/library/HTMLPurifier/DoctypeRegistry.php new file mode 100644 index 000000000..86049e939 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/DoctypeRegistry.php @@ -0,0 +1,103 @@ +<?php + +class HTMLPurifier_DoctypeRegistry +{ + + /** + * Hash of doctype names to doctype objects + */ + protected $doctypes; + + /** + * Lookup table of aliases to real doctype names + */ + protected $aliases; + + /** + * Registers a doctype to the registry + * @note Accepts a fully-formed doctype object, or the + * parameters for constructing a doctype object + * @param $doctype Name of doctype or literal doctype object + * @param $modules Modules doctype will load + * @param $modules_for_modes Modules doctype will load for certain modes + * @param $aliases Alias names for doctype + * @return Editable registered doctype + */ + public function register($doctype, $xml = true, $modules = array(), + $tidy_modules = array(), $aliases = array(), $dtd_public = null, $dtd_system = null + ) { + if (!is_array($modules)) $modules = array($modules); + if (!is_array($tidy_modules)) $tidy_modules = array($tidy_modules); + if (!is_array($aliases)) $aliases = array($aliases); + if (!is_object($doctype)) { + $doctype = new HTMLPurifier_Doctype( + $doctype, $xml, $modules, $tidy_modules, $aliases, $dtd_public, $dtd_system + ); + } + $this->doctypes[$doctype->name] = $doctype; + $name = $doctype->name; + // hookup aliases + foreach ($doctype->aliases as $alias) { + if (isset($this->doctypes[$alias])) continue; + $this->aliases[$alias] = $name; + } + // remove old aliases + if (isset($this->aliases[$name])) unset($this->aliases[$name]); + return $doctype; + } + + /** + * Retrieves reference to a doctype of a certain name + * @note This function resolves aliases + * @note When possible, use the more fully-featured make() + * @param $doctype Name of doctype + * @return Editable doctype object + */ + public function get($doctype) { + if (isset($this->aliases[$doctype])) $doctype = $this->aliases[$doctype]; + if (!isset($this->doctypes[$doctype])) { + trigger_error('Doctype ' . htmlspecialchars($doctype) . ' does not exist', E_USER_ERROR); + $anon = new HTMLPurifier_Doctype($doctype); + return $anon; + } + return $this->doctypes[$doctype]; + } + + /** + * Creates a doctype based on a configuration object, + * will perform initialization on the doctype + * @note Use this function to get a copy of doctype that config + * can hold on to (this is necessary in order to tell + * Generator whether or not the current document is XML + * based or not). + */ + public function make($config) { + return clone $this->get($this->getDoctypeFromConfig($config)); + } + + /** + * Retrieves the doctype from the configuration object + */ + public function getDoctypeFromConfig($config) { + // recommended test + $doctype = $config->get('HTML.Doctype'); + if (!empty($doctype)) return $doctype; + $doctype = $config->get('HTML.CustomDoctype'); + if (!empty($doctype)) return $doctype; + // backwards-compatibility + if ($config->get('HTML.XHTML')) { + $doctype = 'XHTML 1.0'; + } else { + $doctype = 'HTML 4.01'; + } + if ($config->get('HTML.Strict')) { + $doctype .= ' Strict'; + } else { + $doctype .= ' Transitional'; + } + return $doctype; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ElementDef.php b/lib/htmlpurifier/library/HTMLPurifier/ElementDef.php new file mode 100644 index 000000000..5498d9567 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ElementDef.php @@ -0,0 +1,183 @@ +<?php + +/** + * Structure that stores an HTML element definition. Used by + * HTMLPurifier_HTMLDefinition and HTMLPurifier_HTMLModule. + * @note This class is inspected by HTMLPurifier_Printer_HTMLDefinition. + * Please update that class too. + * @warning If you add new properties to this class, you MUST update + * the mergeIn() method. + */ +class HTMLPurifier_ElementDef +{ + + /** + * Does the definition work by itself, or is it created solely + * for the purpose of merging into another definition? + */ + public $standalone = true; + + /** + * Associative array of attribute name to HTMLPurifier_AttrDef + * @note Before being processed by HTMLPurifier_AttrCollections + * when modules are finalized during + * HTMLPurifier_HTMLDefinition->setup(), this array may also + * contain an array at index 0 that indicates which attribute + * collections to load into the full array. It may also + * contain string indentifiers in lieu of HTMLPurifier_AttrDef, + * see HTMLPurifier_AttrTypes on how they are expanded during + * HTMLPurifier_HTMLDefinition->setup() processing. + */ + public $attr = array(); + + /** + * Indexed list of tag's HTMLPurifier_AttrTransform to be done before validation + */ + public $attr_transform_pre = array(); + + /** + * Indexed list of tag's HTMLPurifier_AttrTransform to be done after validation + */ + public $attr_transform_post = array(); + + /** + * HTMLPurifier_ChildDef of this tag. + */ + public $child; + + /** + * Abstract string representation of internal ChildDef rules. See + * HTMLPurifier_ContentSets for how this is parsed and then transformed + * into an HTMLPurifier_ChildDef. + * @warning This is a temporary variable that is not available after + * being processed by HTMLDefinition + */ + public $content_model; + + /** + * Value of $child->type, used to determine which ChildDef to use, + * used in combination with $content_model. + * @warning This must be lowercase + * @warning This is a temporary variable that is not available after + * being processed by HTMLDefinition + */ + public $content_model_type; + + + + /** + * Does the element have a content model (#PCDATA | Inline)*? This + * is important for chameleon ins and del processing in + * HTMLPurifier_ChildDef_Chameleon. Dynamically set: modules don't + * have to worry about this one. + */ + public $descendants_are_inline = false; + + /** + * List of the names of required attributes this element has. Dynamically + * populated by HTMLPurifier_HTMLDefinition::getElement + */ + public $required_attr = array(); + + /** + * Lookup table of tags excluded from all descendants of this tag. + * @note SGML permits exclusions for all descendants, but this is + * not possible with DTDs or XML Schemas. W3C has elected to + * use complicated compositions of content_models to simulate + * exclusion for children, but we go the simpler, SGML-style + * route of flat-out exclusions, which correctly apply to + * all descendants and not just children. Note that the XHTML + * Modularization Abstract Modules are blithely unaware of such + * distinctions. + */ + public $excludes = array(); + + /** + * This tag is explicitly auto-closed by the following tags. + */ + public $autoclose = array(); + + /** + * If a foreign element is found in this element, test if it is + * allowed by this sub-element; if it is, instead of closing the + * current element, place it inside this element. + */ + public $wrap; + + /** + * Whether or not this is a formatting element affected by the + * "Active Formatting Elements" algorithm. + */ + public $formatting; + + /** + * Low-level factory constructor for creating new standalone element defs + */ + public static function create($content_model, $content_model_type, $attr) { + $def = new HTMLPurifier_ElementDef(); + $def->content_model = $content_model; + $def->content_model_type = $content_model_type; + $def->attr = $attr; + return $def; + } + + /** + * Merges the values of another element definition into this one. + * Values from the new element def take precedence if a value is + * not mergeable. + */ + public function mergeIn($def) { + + // later keys takes precedence + foreach($def->attr as $k => $v) { + if ($k === 0) { + // merge in the includes + // sorry, no way to override an include + foreach ($v as $v2) { + $this->attr[0][] = $v2; + } + continue; + } + if ($v === false) { + if (isset($this->attr[$k])) unset($this->attr[$k]); + continue; + } + $this->attr[$k] = $v; + } + $this->_mergeAssocArray($this->attr_transform_pre, $def->attr_transform_pre); + $this->_mergeAssocArray($this->attr_transform_post, $def->attr_transform_post); + $this->_mergeAssocArray($this->excludes, $def->excludes); + + if(!empty($def->content_model)) { + $this->content_model = + str_replace("#SUPER", $this->content_model, $def->content_model); + $this->child = false; + } + if(!empty($def->content_model_type)) { + $this->content_model_type = $def->content_model_type; + $this->child = false; + } + if(!is_null($def->child)) $this->child = $def->child; + if(!is_null($def->formatting)) $this->formatting = $def->formatting; + if($def->descendants_are_inline) $this->descendants_are_inline = $def->descendants_are_inline; + + } + + /** + * Merges one array into another, removes values which equal false + * @param $a1 Array by reference that is merged into + * @param $a2 Array that merges into $a1 + */ + private function _mergeAssocArray(&$a1, $a2) { + foreach ($a2 as $k => $v) { + if ($v === false) { + if (isset($a1[$k])) unset($a1[$k]); + continue; + } + $a1[$k] = $v; + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Encoder.php b/lib/htmlpurifier/library/HTMLPurifier/Encoder.php new file mode 100644 index 000000000..9fa76bd18 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Encoder.php @@ -0,0 +1,540 @@ +<?php + +/** + * A UTF-8 specific character encoder that handles cleaning and transforming. + * @note All functions in this class should be static. + */ +class HTMLPurifier_Encoder +{ + + /** + * Constructor throws fatal error if you attempt to instantiate class + */ + private function __construct() { + trigger_error('Cannot instantiate encoder, call methods statically', E_USER_ERROR); + } + + /** + * Error-handler that mutes errors, alternative to shut-up operator. + */ + public static function muteErrorHandler() {} + + /** + * iconv wrapper which mutes errors, but doesn't work around bugs. + */ + public static function unsafeIconv($in, $out, $text) { + set_error_handler(array('HTMLPurifier_Encoder', 'muteErrorHandler')); + $r = iconv($in, $out, $text); + restore_error_handler(); + return $r; + } + + /** + * iconv wrapper which mutes errors and works around bugs. + */ + public static function iconv($in, $out, $text, $max_chunk_size = 8000) { + $code = self::testIconvTruncateBug(); + if ($code == self::ICONV_OK) { + return self::unsafeIconv($in, $out, $text); + } elseif ($code == self::ICONV_TRUNCATES) { + // we can only work around this if the input character set + // is utf-8 + if ($in == 'utf-8') { + if ($max_chunk_size < 4) { + trigger_error('max_chunk_size is too small', E_USER_WARNING); + return false; + } + // split into 8000 byte chunks, but be careful to handle + // multibyte boundaries properly + if (($c = strlen($text)) <= $max_chunk_size) { + return self::unsafeIconv($in, $out, $text); + } + $r = ''; + $i = 0; + while (true) { + if ($i + $max_chunk_size >= $c) { + $r .= self::unsafeIconv($in, $out, substr($text, $i)); + break; + } + // wibble the boundary + if (0x80 != (0xC0 & ord($text[$i + $max_chunk_size]))) { + $chunk_size = $max_chunk_size; + } elseif (0x80 != (0xC0 & ord($text[$i + $max_chunk_size - 1]))) { + $chunk_size = $max_chunk_size - 1; + } elseif (0x80 != (0xC0 & ord($text[$i + $max_chunk_size - 2]))) { + $chunk_size = $max_chunk_size - 2; + } elseif (0x80 != (0xC0 & ord($text[$i + $max_chunk_size - 3]))) { + $chunk_size = $max_chunk_size - 3; + } else { + return false; // rather confusing UTF-8... + } + $chunk = substr($text, $i, $chunk_size); // substr doesn't mind overlong lengths + $r .= self::unsafeIconv($in, $out, $chunk); + $i += $chunk_size; + } + return $r; + } else { + return false; + } + } else { + return false; + } + } + + /** + * Cleans a UTF-8 string for well-formedness and SGML validity + * + * It will parse according to UTF-8 and return a valid UTF8 string, with + * non-SGML codepoints excluded. + * + * @note Just for reference, the non-SGML code points are 0 to 31 and + * 127 to 159, inclusive. However, we allow code points 9, 10 + * and 13, which are the tab, line feed and carriage return + * respectively. 128 and above the code points map to multibyte + * UTF-8 representations. + * + * @note Fallback code adapted from utf8ToUnicode by Henri Sivonen and + * hsivonen@iki.fi at <http://iki.fi/hsivonen/php-utf8/> under the + * LGPL license. Notes on what changed are inside, but in general, + * the original code transformed UTF-8 text into an array of integer + * Unicode codepoints. Understandably, transforming that back to + * a string would be somewhat expensive, so the function was modded to + * directly operate on the string. However, this discourages code + * reuse, and the logic enumerated here would be useful for any + * function that needs to be able to understand UTF-8 characters. + * As of right now, only smart lossless character encoding converters + * would need that, and I'm probably not going to implement them. + * Once again, PHP 6 should solve all our problems. + */ + public static function cleanUTF8($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)) { + return $str; + } + + $mState = 0; // cached expected number of octets after the current octet + // until the beginning of the next UTF8 character sequence + $mUcs4 = 0; // cached Unicode character + $mBytes = 1; // cached expected number of octets in the current sequence + + // original code involved an $out that was an array of Unicode + // codepoints. Instead of having to convert back into UTF-8, we've + // decided to directly append valid UTF-8 characters onto a string + // $out once they're done. $char accumulates raw bytes, while $mUcs4 + // turns into the Unicode code point, so there's some redundancy. + + $out = ''; + $char = ''; + + $len = strlen($str); + for($i = 0; $i < $len; $i++) { + $in = ord($str{$i}); + $char .= $str[$i]; // append byte to char + if (0 == $mState) { + // When mState is zero we expect either a US-ASCII character + // or a multi-octet sequence. + if (0 == (0x80 & ($in))) { + // US-ASCII, pass straight through. + if (($in <= 31 || $in == 127) && + !($in == 9 || $in == 13 || $in == 10) // save \r\t\n + ) { + // control characters, remove + } else { + $out .= $char; + } + // reset + $char = ''; + $mBytes = 1; + } elseif (0xC0 == (0xE0 & ($in))) { + // First octet of 2 octet sequence + $mUcs4 = ($in); + $mUcs4 = ($mUcs4 & 0x1F) << 6; + $mState = 1; + $mBytes = 2; + } elseif (0xE0 == (0xF0 & ($in))) { + // First octet of 3 octet sequence + $mUcs4 = ($in); + $mUcs4 = ($mUcs4 & 0x0F) << 12; + $mState = 2; + $mBytes = 3; + } elseif (0xF0 == (0xF8 & ($in))) { + // First octet of 4 octet sequence + $mUcs4 = ($in); + $mUcs4 = ($mUcs4 & 0x07) << 18; + $mState = 3; + $mBytes = 4; + } elseif (0xF8 == (0xFC & ($in))) { + // First octet of 5 octet sequence. + // + // This is illegal because the encoded codepoint must be + // either: + // (a) not the shortest form or + // (b) outside the Unicode range of 0-0x10FFFF. + // Rather than trying to resynchronize, we will carry on + // until the end of the sequence and let the later error + // handling code catch it. + $mUcs4 = ($in); + $mUcs4 = ($mUcs4 & 0x03) << 24; + $mState = 4; + $mBytes = 5; + } elseif (0xFC == (0xFE & ($in))) { + // First octet of 6 octet sequence, see comments for 5 + // octet sequence. + $mUcs4 = ($in); + $mUcs4 = ($mUcs4 & 1) << 30; + $mState = 5; + $mBytes = 6; + } else { + // Current octet is neither in the US-ASCII range nor a + // legal first octet of a multi-octet sequence. + $mState = 0; + $mUcs4 = 0; + $mBytes = 1; + $char = ''; + } + } else { + // When mState is non-zero, we expect a continuation of the + // multi-octet sequence + if (0x80 == (0xC0 & ($in))) { + // Legal continuation. + $shift = ($mState - 1) * 6; + $tmp = $in; + $tmp = ($tmp & 0x0000003F) << $shift; + $mUcs4 |= $tmp; + + if (0 == --$mState) { + // End of the multi-octet sequence. mUcs4 now contains + // the final Unicode codepoint to be output + + // Check for illegal sequences and codepoints. + + // From Unicode 3.1, non-shortest form is illegal + if (((2 == $mBytes) && ($mUcs4 < 0x0080)) || + ((3 == $mBytes) && ($mUcs4 < 0x0800)) || + ((4 == $mBytes) && ($mUcs4 < 0x10000)) || + (4 < $mBytes) || + // From Unicode 3.2, surrogate characters = illegal + (($mUcs4 & 0xFFFFF800) == 0xD800) || + // Codepoints outside the Unicode range are illegal + ($mUcs4 > 0x10FFFF) + ) { + + } elseif (0xFEFF != $mUcs4 && // omit BOM + // check for valid Char unicode codepoints + ( + 0x9 == $mUcs4 || + 0xA == $mUcs4 || + 0xD == $mUcs4 || + (0x20 <= $mUcs4 && 0x7E >= $mUcs4) || + // 7F-9F is not strictly prohibited by XML, + // but it is non-SGML, and thus we don't allow it + (0xA0 <= $mUcs4 && 0xD7FF >= $mUcs4) || + (0x10000 <= $mUcs4 && 0x10FFFF >= $mUcs4) + ) + ) { + $out .= $char; + } + // initialize UTF8 cache (reset) + $mState = 0; + $mUcs4 = 0; + $mBytes = 1; + $char = ''; + } + } else { + // ((0xC0 & (*in) != 0x80) && (mState != 0)) + // Incomplete multi-octet sequence. + // used to result in complete fail, but we'll reset + $mState = 0; + $mUcs4 = 0; + $mBytes = 1; + $char =''; + } + } + } + return $out; + } + + /** + * Translates a Unicode codepoint into its corresponding UTF-8 character. + * @note Based on Feyd's function at + * <http://forums.devnetwork.net/viewtopic.php?p=191404#191404>, + * which is in public domain. + * @note While we're going to do code point parsing anyway, a good + * optimization would be to refuse to translate code points that + * are non-SGML characters. However, this could lead to duplication. + * @note This is very similar to the unichr function in + * maintenance/generate-entity-file.php (although this is superior, + * due to its sanity checks). + */ + + // +----------+----------+----------+----------+ + // | 33222222 | 22221111 | 111111 | | + // | 10987654 | 32109876 | 54321098 | 76543210 | bit + // +----------+----------+----------+----------+ + // | | | | 0xxxxxxx | 1 byte 0x00000000..0x0000007F + // | | | 110yyyyy | 10xxxxxx | 2 byte 0x00000080..0x000007FF + // | | 1110zzzz | 10yyyyyy | 10xxxxxx | 3 byte 0x00000800..0x0000FFFF + // | 11110www | 10wwzzzz | 10yyyyyy | 10xxxxxx | 4 byte 0x00010000..0x0010FFFF + // +----------+----------+----------+----------+ + // | 00000000 | 00011111 | 11111111 | 11111111 | Theoretical upper limit of legal scalars: 2097151 (0x001FFFFF) + // | 00000000 | 00010000 | 11111111 | 11111111 | Defined upper limit of legal scalar codes + // +----------+----------+----------+----------+ + + public static function unichr($code) { + if($code > 1114111 or $code < 0 or + ($code >= 55296 and $code <= 57343) ) { + // bits are set outside the "valid" range as defined + // by UNICODE 4.1.0 + return ''; + } + + $x = $y = $z = $w = 0; + if ($code < 128) { + // regular ASCII character + $x = $code; + } else { + // set up bits for UTF-8 + $x = ($code & 63) | 128; + if ($code < 2048) { + $y = (($code & 2047) >> 6) | 192; + } else { + $y = (($code & 4032) >> 6) | 128; + if($code < 65536) { + $z = (($code >> 12) & 15) | 224; + } else { + $z = (($code >> 12) & 63) | 128; + $w = (($code >> 18) & 7) | 240; + } + } + } + // set up the actual character + $ret = ''; + if($w) $ret .= chr($w); + if($z) $ret .= chr($z); + if($y) $ret .= chr($y); + $ret .= chr($x); + + return $ret; + } + + public static function iconvAvailable() { + static $iconv = null; + if ($iconv === null) { + $iconv = function_exists('iconv') && self::testIconvTruncateBug() != self::ICONV_UNUSABLE; + } + return $iconv; + } + + /** + * Converts a string to UTF-8 based on configuration. + */ + public static function convertToUTF8($str, $config, $context) { + $encoding = $config->get('Core.Encoding'); + if ($encoding === 'utf-8') return $str; + static $iconv = null; + if ($iconv === null) $iconv = self::iconvAvailable(); + if ($iconv && !$config->get('Test.ForceNoIconv')) { + // unaffected by bugs, since UTF-8 support all characters + $str = self::unsafeIconv($encoding, 'utf-8//IGNORE', $str); + if ($str === false) { + // $encoding is not a valid encoding + trigger_error('Invalid encoding ' . $encoding, E_USER_ERROR); + return ''; + } + // If the string is bjorked by Shift_JIS or a similar encoding + // that doesn't support all of ASCII, convert the naughty + // characters to their true byte-wise ASCII/UTF-8 equivalents. + $str = strtr($str, self::testEncodingSupportsASCII($encoding)); + return $str; + } elseif ($encoding === 'iso-8859-1') { + $str = utf8_encode($str); + return $str; + } + trigger_error('Encoding not supported, please install iconv', E_USER_ERROR); + } + + /** + * Converts a string from UTF-8 based on configuration. + * @note Currently, this is a lossy conversion, with unexpressable + * characters being omitted. + */ + public static function convertFromUTF8($str, $config, $context) { + $encoding = $config->get('Core.Encoding'); + if ($escape = $config->get('Core.EscapeNonASCIICharacters')) { + $str = self::convertToASCIIDumbLossless($str); + } + if ($encoding === 'utf-8') return $str; + static $iconv = null; + if ($iconv === null) $iconv = self::iconvAvailable(); + if ($iconv && !$config->get('Test.ForceNoIconv')) { + // Undo our previous fix in convertToUTF8, otherwise iconv will barf + $ascii_fix = self::testEncodingSupportsASCII($encoding); + if (!$escape && !empty($ascii_fix)) { + $clear_fix = array(); + foreach ($ascii_fix as $utf8 => $native) $clear_fix[$utf8] = ''; + $str = strtr($str, $clear_fix); + } + $str = strtr($str, array_flip($ascii_fix)); + // Normal stuff + $str = self::iconv('utf-8', $encoding . '//IGNORE', $str); + return $str; + } elseif ($encoding === 'iso-8859-1') { + $str = utf8_decode($str); + return $str; + } + trigger_error('Encoding not supported', E_USER_ERROR); + // You might be tempted to assume that the ASCII representation + // might be OK, however, this is *not* universally true over all + // encodings. So we take the conservative route here, rather + // than forcibly turn on %Core.EscapeNonASCIICharacters + } + + /** + * Lossless (character-wise) conversion of HTML to ASCII + * @param $str UTF-8 string to be converted to ASCII + * @returns ASCII encoded string with non-ASCII character entity-ized + * @warning Adapted from MediaWiki, claiming fair use: this is a common + * algorithm. If you disagree with this license fudgery, + * implement it yourself. + * @note Uses decimal numeric entities since they are best supported. + * @note This is a DUMB function: it has no concept of keeping + * character entities that the projected character encoding + * can allow. We could possibly implement a smart version + * but that would require it to also know which Unicode + * codepoints the charset supported (not an easy task). + * @note Sort of with cleanUTF8() but it assumes that $str is + * well-formed UTF-8 + */ + public static function convertToASCIIDumbLossless($str) { + $bytesleft = 0; + $result = ''; + $working = 0; + $len = strlen($str); + for( $i = 0; $i < $len; $i++ ) { + $bytevalue = ord( $str[$i] ); + if( $bytevalue <= 0x7F ) { //0xxx xxxx + $result .= chr( $bytevalue ); + $bytesleft = 0; + } elseif( $bytevalue <= 0xBF ) { //10xx xxxx + $working = $working << 6; + $working += ($bytevalue & 0x3F); + $bytesleft--; + if( $bytesleft <= 0 ) { + $result .= "&#" . $working . ";"; + } + } elseif( $bytevalue <= 0xDF ) { //110x xxxx + $working = $bytevalue & 0x1F; + $bytesleft = 1; + } elseif( $bytevalue <= 0xEF ) { //1110 xxxx + $working = $bytevalue & 0x0F; + $bytesleft = 2; + } else { //1111 0xxx + $working = $bytevalue & 0x07; + $bytesleft = 3; + } + } + return $result; + } + + /** No bugs detected in iconv. */ + const ICONV_OK = 0; + + /** Iconv truncates output if converting from UTF-8 to another + * character set with //IGNORE, and a non-encodable character is found */ + const ICONV_TRUNCATES = 1; + + /** Iconv does not support //IGNORE, making it unusable for + * transcoding purposes */ + const ICONV_UNUSABLE = 2; + + /** + * glibc iconv has a known bug where it doesn't handle the magic + * //IGNORE stanza correctly. In particular, rather than ignore + * characters, it will return an EILSEQ after consuming some number + * of characters, and expect you to restart iconv as if it were + * an E2BIG. Old versions of PHP did not respect the errno, and + * returned the fragment, so as a result you would see iconv + * mysteriously truncating output. We can work around this by + * manually chopping our input into segments of about 8000 + * characters, as long as PHP ignores the error code. If PHP starts + * paying attention to the error code, iconv becomes unusable. + * + * @returns Error code indicating severity of bug. + */ + public static function testIconvTruncateBug() { + static $code = null; + if ($code === null) { + // better not use iconv, otherwise infinite loop! + $r = self::unsafeIconv('utf-8', 'ascii//IGNORE', "\xCE\xB1" . str_repeat('a', 9000)); + if ($r === false) { + $code = self::ICONV_UNUSABLE; + } elseif (($c = strlen($r)) < 9000) { + $code = self::ICONV_TRUNCATES; + } elseif ($c > 9000) { + trigger_error('Your copy of iconv is extremely buggy. Please notify HTML Purifier maintainers: include your iconv version as per phpversion()', E_USER_ERROR); + } else { + $code = self::ICONV_OK; + } + } + return $code; + } + + /** + * This expensive function tests whether or not a given character + * encoding supports ASCII. 7/8-bit encodings like Shift_JIS will + * fail this test, and require special processing. Variable width + * encodings shouldn't ever fail. + * + * @param string $encoding Encoding name to test, as per iconv format + * @param bool $bypass Whether or not to bypass the precompiled arrays. + * @return Array of UTF-8 characters to their corresponding ASCII, + * which can be used to "undo" any overzealous iconv action. + */ + public static function testEncodingSupportsASCII($encoding, $bypass = false) { + // All calls to iconv here are unsafe, proof by case analysis: + // If ICONV_OK, no difference. + // If ICONV_TRUNCATE, all calls involve one character inputs, + // so bug is not triggered. + // If ICONV_UNUSABLE, this call is irrelevant + static $encodings = array(); + if (!$bypass) { + if (isset($encodings[$encoding])) return $encodings[$encoding]; + $lenc = strtolower($encoding); + switch ($lenc) { + case 'shift_jis': + return array("\xC2\xA5" => '\\', "\xE2\x80\xBE" => '~'); + case 'johab': + return array("\xE2\x82\xA9" => '\\'); + } + if (strpos($lenc, 'iso-8859-') === 0) return array(); + } + $ret = array(); + if (self::unsafeIconv('UTF-8', $encoding, 'a') === false) return false; + for ($i = 0x20; $i <= 0x7E; $i++) { // all printable ASCII chars + $c = chr($i); // UTF-8 char + $r = self::unsafeIconv('UTF-8', "$encoding//IGNORE", $c); // initial conversion + if ( + $r === '' || + // This line is needed for iconv implementations that do not + // omit characters that do not exist in the target character set + ($r === $c && self::unsafeIconv($encoding, 'UTF-8//IGNORE', $r) !== $c) + ) { + // Reverse engineer: what's the UTF-8 equiv of this byte + // sequence? This assumes that there's no variable width + // encoding that doesn't support ASCII. + $ret[self::unsafeIconv($encoding, 'UTF-8//IGNORE', $c)] = $c; + } + } + $encodings[$encoding] = $ret; + return $ret; + } + + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/EntityLookup.php b/lib/htmlpurifier/library/HTMLPurifier/EntityLookup.php new file mode 100644 index 000000000..b4dfce94c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/EntityLookup.php @@ -0,0 +1,44 @@ +<?php + +/** + * Object that provides entity lookup table from entity name to character + */ +class HTMLPurifier_EntityLookup { + + /** + * Assoc array of entity name to character represented. + */ + public $table; + + /** + * Sets up the entity lookup table from the serialized file contents. + * @note The serialized contents are versioned, but were generated + * using the maintenance script generate_entity_file.php + * @warning This is not in constructor to help enforce the Singleton + */ + public function setup($file = false) { + if (!$file) { + $file = HTMLPURIFIER_PREFIX . '/HTMLPurifier/EntityLookup/entities.ser'; + } + $this->table = unserialize(file_get_contents($file)); + } + + /** + * Retrieves sole instance of the object. + * @param Optional prototype of custom lookup table to overload with. + */ + public static function instance($prototype = false) { + // no references, since PHP doesn't copy unless modified + static $instance = null; + if ($prototype) { + $instance = $prototype; + } elseif (!$instance) { + $instance = new HTMLPurifier_EntityLookup(); + $instance->setup(); + } + return $instance; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/EntityLookup/entities.ser b/lib/htmlpurifier/library/HTMLPurifier/EntityLookup/entities.ser new file mode 100644 index 000000000..e8b08128b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/EntityLookup/entities.ser @@ -0,0 +1 @@ +a:253:{s:4:"fnof";s:2:"ƒ";s:5:"Alpha";s:2:"Α";s:4:"Beta";s:2:"Β";s:5:"Gamma";s:2:"Γ";s:5:"Delta";s:2:"Δ";s:7:"Epsilon";s:2:"Ε";s:4:"Zeta";s:2:"Ζ";s:3:"Eta";s:2:"Η";s:5:"Theta";s:2:"Θ";s:4:"Iota";s:2:"Ι";s:5:"Kappa";s:2:"Κ";s:6:"Lambda";s:2:"Λ";s:2:"Mu";s:2:"Μ";s:2:"Nu";s:2:"Ν";s:2:"Xi";s:2:"Ξ";s:7:"Omicron";s:2:"Ο";s:2:"Pi";s:2:"Π";s:3:"Rho";s:2:"Ρ";s:5:"Sigma";s:2:"Σ";s:3:"Tau";s:2:"Τ";s:7:"Upsilon";s:2:"Υ";s:3:"Phi";s:2:"Φ";s:3:"Chi";s:2:"Χ";s:3:"Psi";s:2:"Ψ";s:5:"Omega";s:2:"Ω";s:5:"alpha";s:2:"α";s:4:"beta";s:2:"β";s:5:"gamma";s:2:"γ";s:5:"delta";s:2:"δ";s:7:"epsilon";s:2:"ε";s:4:"zeta";s:2:"ζ";s:3:"eta";s:2:"η";s:5:"theta";s:2:"θ";s:4:"iota";s:2:"ι";s:5:"kappa";s:2:"κ";s:6:"lambda";s:2:"λ";s:2:"mu";s:2:"μ";s:2:"nu";s:2:"ν";s:2:"xi";s:2:"ξ";s:7:"omicron";s:2:"ο";s:2:"pi";s:2:"π";s:3:"rho";s:2:"ρ";s:6:"sigmaf";s:2:"ς";s:5:"sigma";s:2:"σ";s:3:"tau";s:2:"τ";s:7:"upsilon";s:2:"υ";s:3:"phi";s:2:"φ";s:3:"chi";s:2:"χ";s:3:"psi";s:2:"ψ";s:5:"omega";s:2:"ω";s:8:"thetasym";s:2:"ϑ";s:5:"upsih";s:2:"ϒ";s:3:"piv";s:2:"ϖ";s:4:"bull";s:3:"•";s:6:"hellip";s:3:"…";s:5:"prime";s:3:"′";s:5:"Prime";s:3:"″";s:5:"oline";s:3:"‾";s:5:"frasl";s:3:"⁄";s:6:"weierp";s:3:"℘";s:5:"image";s:3:"ℑ";s:4:"real";s:3:"ℜ";s:5:"trade";s:3:"™";s:7:"alefsym";s:3:"ℵ";s:4:"larr";s:3:"←";s:4:"uarr";s:3:"↑";s:4:"rarr";s:3:"→";s:4:"darr";s:3:"↓";s:4:"harr";s:3:"↔";s:5:"crarr";s:3:"↵";s:4:"lArr";s:3:"⇐";s:4:"uArr";s:3:"⇑";s:4:"rArr";s:3:"⇒";s:4:"dArr";s:3:"⇓";s:4:"hArr";s:3:"⇔";s:6:"forall";s:3:"∀";s:4:"part";s:3:"∂";s:5:"exist";s:3:"∃";s:5:"empty";s:3:"∅";s:5:"nabla";s:3:"∇";s:4:"isin";s:3:"∈";s:5:"notin";s:3:"∉";s:2:"ni";s:3:"∋";s:4:"prod";s:3:"∏";s:3:"sum";s:3:"∑";s:5:"minus";s:3:"−";s:6:"lowast";s:3:"∗";s:5:"radic";s:3:"√";s:4:"prop";s:3:"∝";s:5:"infin";s:3:"∞";s:3:"ang";s:3:"∠";s:3:"and";s:3:"∧";s:2:"or";s:3:"∨";s:3:"cap";s:3:"∩";s:3:"cup";s:3:"∪";s:3:"int";s:3:"∫";s:6:"there4";s:3:"∴";s:3:"sim";s:3:"∼";s:4:"cong";s:3:"≅";s:5:"asymp";s:3:"≈";s:2:"ne";s:3:"≠";s:5:"equiv";s:3:"≡";s:2:"le";s:3:"≤";s:2:"ge";s:3:"≥";s:3:"sub";s:3:"⊂";s:3:"sup";s:3:"⊃";s:4:"nsub";s:3:"⊄";s:4:"sube";s:3:"⊆";s:4:"supe";s:3:"⊇";s:5:"oplus";s:3:"⊕";s:6:"otimes";s:3:"⊗";s:4:"perp";s:3:"⊥";s:4:"sdot";s:3:"⋅";s:5:"lceil";s:3:"⌈";s:5:"rceil";s:3:"⌉";s:6:"lfloor";s:3:"⌊";s:6:"rfloor";s:3:"⌋";s:4:"lang";s:3:"〈";s:4:"rang";s:3:"〉";s:3:"loz";s:3:"◊";s:6:"spades";s:3:"♠";s:5:"clubs";s:3:"♣";s:6:"hearts";s:3:"♥";s:5:"diams";s:3:"♦";s:4:"quot";s:1:""";s:3:"amp";s:1:"&";s:2:"lt";s:1:"<";s:2:"gt";s:1:">";s:4:"apos";s:1:"'";s:5:"OElig";s:2:"Œ";s:5:"oelig";s:2:"œ";s:6:"Scaron";s:2:"Š";s:6:"scaron";s:2:"š";s:4:"Yuml";s:2:"Ÿ";s:4:"circ";s:2:"ˆ";s:5:"tilde";s:2:"˜";s:4:"ensp";s:3:" ";s:4:"emsp";s:3:" ";s:6:"thinsp";s:3:" ";s:4:"zwnj";s:3:"";s:3:"zwj";s:3:"";s:3:"lrm";s:3:"";s:3:"rlm";s:3:"";s:5:"ndash";s:3:"–";s:5:"mdash";s:3:"—";s:5:"lsquo";s:3:"‘";s:5:"rsquo";s:3:"’";s:5:"sbquo";s:3:"‚";s:5:"ldquo";s:3:"“";s:5:"rdquo";s:3:"”";s:5:"bdquo";s:3:"„";s:6:"dagger";s:3:"†";s:6:"Dagger";s:3:"‡";s:6:"permil";s:3:"‰";s:6:"lsaquo";s:3:"‹";s:6:"rsaquo";s:3:"›";s:4:"euro";s:3:"€";s:4:"nbsp";s:2:" ";s:5:"iexcl";s:2:"¡";s:4:"cent";s:2:"¢";s:5:"pound";s:2:"£";s:6:"curren";s:2:"¤";s:3:"yen";s:2:"¥";s:6:"brvbar";s:2:"¦";s:4:"sect";s:2:"§";s:3:"uml";s:2:"¨";s:4:"copy";s:2:"©";s:4:"ordf";s:2:"ª";s:5:"laquo";s:2:"«";s:3:"not";s:2:"¬";s:3:"shy";s:2:"";s:3:"reg";s:2:"®";s:4:"macr";s:2:"¯";s:3:"deg";s:2:"°";s:6:"plusmn";s:2:"±";s:4:"sup2";s:2:"²";s:4:"sup3";s:2:"³";s:5:"acute";s:2:"´";s:5:"micro";s:2:"µ";s:4:"para";s:2:"¶";s:6:"middot";s:2:"·";s:5:"cedil";s:2:"¸";s:4:"sup1";s:2:"¹";s:4:"ordm";s:2:"º";s:5:"raquo";s:2:"»";s:6:"frac14";s:2:"¼";s:6:"frac12";s:2:"½";s:6:"frac34";s:2:"¾";s:6:"iquest";s:2:"¿";s:6:"Agrave";s:2:"À";s:6:"Aacute";s:2:"Á";s:5:"Acirc";s:2:"Â";s:6:"Atilde";s:2:"Ã";s:4:"Auml";s:2:"Ä";s:5:"Aring";s:2:"Å";s:5:"AElig";s:2:"Æ";s:6:"Ccedil";s:2:"Ç";s:6:"Egrave";s:2:"È";s:6:"Eacute";s:2:"É";s:5:"Ecirc";s:2:"Ê";s:4:"Euml";s:2:"Ë";s:6:"Igrave";s:2:"Ì";s:6:"Iacute";s:2:"Í";s:5:"Icirc";s:2:"Î";s:4:"Iuml";s:2:"Ï";s:3:"ETH";s:2:"Ð";s:6:"Ntilde";s:2:"Ñ";s:6:"Ograve";s:2:"Ò";s:6:"Oacute";s:2:"Ó";s:5:"Ocirc";s:2:"Ô";s:6:"Otilde";s:2:"Õ";s:4:"Ouml";s:2:"Ö";s:5:"times";s:2:"×";s:6:"Oslash";s:2:"Ø";s:6:"Ugrave";s:2:"Ù";s:6:"Uacute";s:2:"Ú";s:5:"Ucirc";s:2:"Û";s:4:"Uuml";s:2:"Ü";s:6:"Yacute";s:2:"Ý";s:5:"THORN";s:2:"Þ";s:5:"szlig";s:2:"ß";s:6:"agrave";s:2:"à";s:6:"aacute";s:2:"á";s:5:"acirc";s:2:"â";s:6:"atilde";s:2:"ã";s:4:"auml";s:2:"ä";s:5:"aring";s:2:"å";s:5:"aelig";s:2:"æ";s:6:"ccedil";s:2:"ç";s:6:"egrave";s:2:"è";s:6:"eacute";s:2:"é";s:5:"ecirc";s:2:"ê";s:4:"euml";s:2:"ë";s:6:"igrave";s:2:"ì";s:6:"iacute";s:2:"í";s:5:"icirc";s:2:"î";s:4:"iuml";s:2:"ï";s:3:"eth";s:2:"ð";s:6:"ntilde";s:2:"ñ";s:6:"ograve";s:2:"ò";s:6:"oacute";s:2:"ó";s:5:"ocirc";s:2:"ô";s:6:"otilde";s:2:"õ";s:4:"ouml";s:2:"ö";s:6:"divide";s:2:"÷";s:6:"oslash";s:2:"ø";s:6:"ugrave";s:2:"ù";s:6:"uacute";s:2:"ú";s:5:"ucirc";s:2:"û";s:4:"uuml";s:2:"ü";s:6:"yacute";s:2:"ý";s:5:"thorn";s:2:"þ";s:4:"yuml";s:2:"ÿ";}
\ No newline at end of file diff --git a/lib/htmlpurifier/library/HTMLPurifier/EntityParser.php b/lib/htmlpurifier/library/HTMLPurifier/EntityParser.php new file mode 100644 index 000000000..8c384472d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/EntityParser.php @@ -0,0 +1,144 @@ +<?php + +// if want to implement error collecting here, we'll need to use some sort +// of global data (probably trigger_error) because it's impossible to pass +// $config or $context to the callback functions. + +/** + * Handles referencing and derefencing character entities + */ +class HTMLPurifier_EntityParser +{ + + /** + * Reference to entity lookup table. + */ + protected $_entity_lookup; + + /** + * Callback regex string for parsing entities. + */ + protected $_substituteEntitiesRegex = +'/&(?:[#]x([a-fA-F0-9]+)|[#]0*(\d+)|([A-Za-z_:][A-Za-z0-9.\-_:]*));?/'; +// 1. hex 2. dec 3. string (XML style) + + + /** + * Decimal to parsed string conversion table for special entities. + */ + protected $_special_dec2str = + array( + 34 => '"', + 38 => '&', + 39 => "'", + 60 => '<', + 62 => '>' + ); + + /** + * Stripped entity names to decimal conversion table for special entities. + */ + protected $_special_ent2dec = + array( + 'quot' => 34, + 'amp' => 38, + 'lt' => 60, + 'gt' => 62 + ); + + /** + * Substitutes non-special entities with their parsed equivalents. Since + * running this whenever you have parsed character is t3h 5uck, we run + * it before everything else. + * + * @param $string String to have non-special entities parsed. + * @returns Parsed string. + */ + public function substituteNonSpecialEntities($string) { + // it will try to detect missing semicolons, but don't rely on it + return preg_replace_callback( + $this->_substituteEntitiesRegex, + array($this, 'nonSpecialEntityCallback'), + $string + ); + } + + /** + * Callback function for substituteNonSpecialEntities() that does the work. + * + * @param $matches PCRE matches array, with 0 the entire match, and + * either index 1, 2 or 3 set with a hex value, dec value, + * or string (respectively). + * @returns Replacement string. + */ + + protected function nonSpecialEntityCallback($matches) { + // replaces all but big five + $entity = $matches[0]; + $is_num = (@$matches[0][1] === '#'); + if ($is_num) { + $is_hex = (@$entity[2] === 'x'); + $code = $is_hex ? hexdec($matches[1]) : (int) $matches[2]; + + // abort for special characters + if (isset($this->_special_dec2str[$code])) return $entity; + + return HTMLPurifier_Encoder::unichr($code); + } else { + if (isset($this->_special_ent2dec[$matches[3]])) return $entity; + if (!$this->_entity_lookup) { + $this->_entity_lookup = HTMLPurifier_EntityLookup::instance(); + } + if (isset($this->_entity_lookup->table[$matches[3]])) { + return $this->_entity_lookup->table[$matches[3]]; + } else { + return $entity; + } + } + } + + /** + * Substitutes only special entities with their parsed equivalents. + * + * @notice We try to avoid calling this function because otherwise, it + * would have to be called a lot (for every parsed section). + * + * @param $string String to have non-special entities parsed. + * @returns Parsed string. + */ + public function substituteSpecialEntities($string) { + return preg_replace_callback( + $this->_substituteEntitiesRegex, + array($this, 'specialEntityCallback'), + $string); + } + + /** + * Callback function for substituteSpecialEntities() that does the work. + * + * This callback has same syntax as nonSpecialEntityCallback(). + * + * @param $matches PCRE-style matches array, with 0 the entire match, and + * either index 1, 2 or 3 set with a hex value, dec value, + * or string (respectively). + * @returns Replacement string. + */ + protected function specialEntityCallback($matches) { + $entity = $matches[0]; + $is_num = (@$matches[0][1] === '#'); + if ($is_num) { + $is_hex = (@$entity[2] === 'x'); + $int = $is_hex ? hexdec($matches[1]) : (int) $matches[2]; + return isset($this->_special_dec2str[$int]) ? + $this->_special_dec2str[$int] : + $entity; + } else { + return isset($this->_special_ent2dec[$matches[3]]) ? + $this->_special_ent2dec[$matches[3]] : + $entity; + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ErrorCollector.php b/lib/htmlpurifier/library/HTMLPurifier/ErrorCollector.php new file mode 100644 index 000000000..6713eaf77 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ErrorCollector.php @@ -0,0 +1,209 @@ +<?php + +/** + * Error collection class that enables HTML Purifier to report HTML + * problems back to the user + */ +class HTMLPurifier_ErrorCollector +{ + + /** + * Identifiers for the returned error array. These are purposely numeric + * so list() can be used. + */ + const LINENO = 0; + const SEVERITY = 1; + const MESSAGE = 2; + const CHILDREN = 3; + + protected $errors; + protected $_current; + protected $_stacks = array(array()); + protected $locale; + protected $generator; + protected $context; + + protected $lines = array(); + + public function __construct($context) { + $this->locale =& $context->get('Locale'); + $this->context = $context; + $this->_current =& $this->_stacks[0]; + $this->errors =& $this->_stacks[0]; + } + + /** + * Sends an error message to the collector for later use + * @param $severity int Error severity, PHP error style (don't use E_USER_) + * @param $msg string Error message text + * @param $subst1 string First substitution for $msg + * @param $subst2 string ... + */ + public function send($severity, $msg) { + + $args = array(); + if (func_num_args() > 2) { + $args = func_get_args(); + array_shift($args); + unset($args[0]); + } + + $token = $this->context->get('CurrentToken', true); + $line = $token ? $token->line : $this->context->get('CurrentLine', true); + $col = $token ? $token->col : $this->context->get('CurrentCol', true); + $attr = $this->context->get('CurrentAttr', true); + + // perform special substitutions, also add custom parameters + $subst = array(); + if (!is_null($token)) { + $args['CurrentToken'] = $token; + } + if (!is_null($attr)) { + $subst['$CurrentAttr.Name'] = $attr; + if (isset($token->attr[$attr])) $subst['$CurrentAttr.Value'] = $token->attr[$attr]; + } + + if (empty($args)) { + $msg = $this->locale->getMessage($msg); + } else { + $msg = $this->locale->formatMessage($msg, $args); + } + + if (!empty($subst)) $msg = strtr($msg, $subst); + + // (numerically indexed) + $error = array( + self::LINENO => $line, + self::SEVERITY => $severity, + self::MESSAGE => $msg, + self::CHILDREN => array() + ); + $this->_current[] = $error; + + + // NEW CODE BELOW ... + + $struct = null; + // Top-level errors are either: + // TOKEN type, if $value is set appropriately, or + // "syntax" type, if $value is null + $new_struct = new HTMLPurifier_ErrorStruct(); + $new_struct->type = HTMLPurifier_ErrorStruct::TOKEN; + if ($token) $new_struct->value = clone $token; + if (is_int($line) && is_int($col)) { + if (isset($this->lines[$line][$col])) { + $struct = $this->lines[$line][$col]; + } else { + $struct = $this->lines[$line][$col] = $new_struct; + } + // These ksorts may present a performance problem + ksort($this->lines[$line], SORT_NUMERIC); + } else { + if (isset($this->lines[-1])) { + $struct = $this->lines[-1]; + } else { + $struct = $this->lines[-1] = $new_struct; + } + } + ksort($this->lines, SORT_NUMERIC); + + // Now, check if we need to operate on a lower structure + if (!empty($attr)) { + $struct = $struct->getChild(HTMLPurifier_ErrorStruct::ATTR, $attr); + if (!$struct->value) { + $struct->value = array($attr, 'PUT VALUE HERE'); + } + } + if (!empty($cssprop)) { + $struct = $struct->getChild(HTMLPurifier_ErrorStruct::CSSPROP, $cssprop); + if (!$struct->value) { + // if we tokenize CSS this might be a little more difficult to do + $struct->value = array($cssprop, 'PUT VALUE HERE'); + } + } + + // Ok, structs are all setup, now time to register the error + $struct->addError($severity, $msg); + } + + /** + * Retrieves raw error data for custom formatter to use + * @param List of arrays in format of array(line of error, + * error severity, error message, + * recursive sub-errors array) + */ + public function getRaw() { + return $this->errors; + } + + /** + * Default HTML formatting implementation for error messages + * @param $config Configuration array, vital for HTML output nature + * @param $errors Errors array to display; used for recursion. + */ + public function getHTMLFormatted($config, $errors = null) { + $ret = array(); + + $this->generator = new HTMLPurifier_Generator($config, $this->context); + if ($errors === null) $errors = $this->errors; + + // 'At line' message needs to be removed + + // generation code for new structure goes here. It needs to be recursive. + foreach ($this->lines as $line => $col_array) { + if ($line == -1) continue; + foreach ($col_array as $col => $struct) { + $this->_renderStruct($ret, $struct, $line, $col); + } + } + if (isset($this->lines[-1])) { + $this->_renderStruct($ret, $this->lines[-1]); + } + + if (empty($errors)) { + return '<p>' . $this->locale->getMessage('ErrorCollector: No errors') . '</p>'; + } else { + return '<ul><li>' . implode('</li><li>', $ret) . '</li></ul>'; + } + + } + + private function _renderStruct(&$ret, $struct, $line = null, $col = null) { + $stack = array($struct); + $context_stack = array(array()); + while ($current = array_pop($stack)) { + $context = array_pop($context_stack); + foreach ($current->errors as $error) { + list($severity, $msg) = $error; + $string = ''; + $string .= '<div>'; + // W3C uses an icon to indicate the severity of the error. + $error = $this->locale->getErrorName($severity); + $string .= "<span class=\"error e$severity\"><strong>$error</strong></span> "; + if (!is_null($line) && !is_null($col)) { + $string .= "<em class=\"location\">Line $line, Column $col: </em> "; + } else { + $string .= '<em class="location">End of Document: </em> '; + } + $string .= '<strong class="description">' . $this->generator->escape($msg) . '</strong> '; + $string .= '</div>'; + // Here, have a marker for the character on the column appropriate. + // Be sure to clip extremely long lines. + //$string .= '<pre>'; + //$string .= ''; + //$string .= '</pre>'; + $ret[] = $string; + } + foreach ($current->children as $type => $array) { + $context[] = $current; + $stack = array_merge($stack, array_reverse($array, true)); + for ($i = count($array); $i > 0; $i--) { + $context_stack[] = $context; + } + } + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/ErrorStruct.php b/lib/htmlpurifier/library/HTMLPurifier/ErrorStruct.php new file mode 100644 index 000000000..9bc8996ec --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/ErrorStruct.php @@ -0,0 +1,60 @@ +<?php + +/** + * Records errors for particular segments of an HTML document such as tokens, + * attributes or CSS properties. They can contain error structs (which apply + * to components of what they represent), but their main purpose is to hold + * errors applying to whatever struct is being used. + */ +class HTMLPurifier_ErrorStruct +{ + + /** + * Possible values for $children first-key. Note that top-level structures + * are automatically token-level. + */ + const TOKEN = 0; + const ATTR = 1; + const CSSPROP = 2; + + /** + * Type of this struct. + */ + public $type; + + /** + * Value of the struct we are recording errors for. There are various + * values for this: + * - TOKEN: Instance of HTMLPurifier_Token + * - ATTR: array('attr-name', 'value') + * - CSSPROP: array('prop-name', 'value') + */ + public $value; + + /** + * Errors registered for this structure. + */ + public $errors = array(); + + /** + * Child ErrorStructs that are from this structure. For example, a TOKEN + * ErrorStruct would contain ATTR ErrorStructs. This is a multi-dimensional + * array in structure: [TYPE]['identifier'] + */ + public $children = array(); + + public function getChild($type, $id) { + if (!isset($this->children[$type][$id])) { + $this->children[$type][$id] = new HTMLPurifier_ErrorStruct(); + $this->children[$type][$id]->type = $type; + } + return $this->children[$type][$id]; + } + + public function addError($severity, $message) { + $this->errors[] = array($severity, $message); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Exception.php b/lib/htmlpurifier/library/HTMLPurifier/Exception.php new file mode 100644 index 000000000..be85b4c56 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Exception.php @@ -0,0 +1,12 @@ +<?php + +/** + * Global exception class for HTML Purifier; any exceptions we throw + * are from here. + */ +class HTMLPurifier_Exception extends Exception +{ + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Filter.php b/lib/htmlpurifier/library/HTMLPurifier/Filter.php new file mode 100644 index 000000000..9a0e7b09f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Filter.php @@ -0,0 +1,46 @@ +<?php + +/** + * Represents a pre or post processing filter on HTML Purifier's output + * + * Sometimes, a little ad-hoc fixing of HTML has to be done before + * it gets sent through HTML Purifier: you can use filters to acheive + * this effect. For instance, YouTube videos can be preserved using + * this manner. You could have used a decorator for this task, but + * PHP's support for them is not terribly robust, so we're going + * to just loop through the filters. + * + * Filters should be exited first in, last out. If there are three filters, + * named 1, 2 and 3, the order of execution should go 1->preFilter, + * 2->preFilter, 3->preFilter, purify, 3->postFilter, 2->postFilter, + * 1->postFilter. + * + * @note Methods are not declared abstract as it is perfectly legitimate + * for an implementation not to want anything to happen on a step + */ + +class HTMLPurifier_Filter +{ + + /** + * Name of the filter for identification purposes + */ + public $name; + + /** + * Pre-processor function, handles HTML before HTML Purifier + */ + public function preFilter($html, $config, $context) { + return $html; + } + + /** + * Post-processor function, handles HTML after HTML Purifier + */ + public function postFilter($html, $config, $context) { + return $html; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Filter/ExtractStyleBlocks.php b/lib/htmlpurifier/library/HTMLPurifier/Filter/ExtractStyleBlocks.php new file mode 100644 index 000000000..320aa4f16 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Filter/ExtractStyleBlocks.php @@ -0,0 +1,288 @@ +<?php + +// why is this a top level function? Because PHP 5.2.0 doesn't seem to +// understand how to interpret this filter if it's a static method. +// It's all really silly, but if we go this route it might be reasonable +// to coalesce all of these methods into one. +function htmlpurifier_filter_extractstyleblocks_muteerrorhandler() {} + +/** + * This filter extracts <style> blocks from input HTML, cleans them up + * using CSSTidy, and then places them in $purifier->context->get('StyleBlocks') + * so they can be used elsewhere in the document. + * + * @note + * See tests/HTMLPurifier/Filter/ExtractStyleBlocksTest.php for + * sample usage. + * + * @note + * This filter can also be used on stylesheets not included in the + * document--something purists would probably prefer. Just directly + * call HTMLPurifier_Filter_ExtractStyleBlocks->cleanCSS() + */ +class HTMLPurifier_Filter_ExtractStyleBlocks extends HTMLPurifier_Filter +{ + + public $name = 'ExtractStyleBlocks'; + private $_styleMatches = array(); + private $_tidy; + + private $_id_attrdef; + private $_class_attrdef; + private $_enum_attrdef; + + public function __construct() { + $this->_tidy = new csstidy(); + $this->_id_attrdef = new HTMLPurifier_AttrDef_HTML_ID(true); + $this->_class_attrdef = new HTMLPurifier_AttrDef_CSS_Ident(); + $this->_enum_attrdef = new HTMLPurifier_AttrDef_Enum(array('first-child', 'link', 'visited', 'active', 'hover', 'focus')); + } + + /** + * Save the contents of CSS blocks to style matches + * @param $matches preg_replace style $matches array + */ + protected function styleCallback($matches) { + $this->_styleMatches[] = $matches[1]; + } + + /** + * Removes inline <style> tags from HTML, saves them for later use + * @todo Extend to indicate non-text/css style blocks + */ + public function preFilter($html, $config, $context) { + $tidy = $config->get('Filter.ExtractStyleBlocks.TidyImpl'); + if ($tidy !== null) $this->_tidy = $tidy; + $html = preg_replace_callback('#<style(?:\s.*)?>(.+)</style>#isU', array($this, 'styleCallback'), $html); + $style_blocks = $this->_styleMatches; + $this->_styleMatches = array(); // reset + $context->register('StyleBlocks', $style_blocks); // $context must not be reused + if ($this->_tidy) { + foreach ($style_blocks as &$style) { + $style = $this->cleanCSS($style, $config, $context); + } + } + return $html; + } + + /** + * Takes CSS (the stuff found in <style>) and cleans it. + * @warning Requires CSSTidy <http://csstidy.sourceforge.net/> + * @param $css CSS styling to clean + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return Cleaned CSS + */ + public function cleanCSS($css, $config, $context) { + // prepare scope + $scope = $config->get('Filter.ExtractStyleBlocks.Scope'); + if ($scope !== null) { + $scopes = array_map('trim', explode(',', $scope)); + } else { + $scopes = array(); + } + // remove comments from CSS + $css = trim($css); + if (strncmp('<!--', $css, 4) === 0) { + $css = substr($css, 4); + } + if (strlen($css) > 3 && substr($css, -3) == '-->') { + $css = substr($css, 0, -3); + } + $css = trim($css); + set_error_handler('htmlpurifier_filter_extractstyleblocks_muteerrorhandler'); + $this->_tidy->parse($css); + restore_error_handler(); + $css_definition = $config->getDefinition('CSS'); + $html_definition = $config->getDefinition('HTML'); + $new_css = array(); + foreach ($this->_tidy->css as $k => $decls) { + // $decls are all CSS declarations inside an @ selector + $new_decls = array(); + foreach ($decls as $selector => $style) { + $selector = trim($selector); + if ($selector === '') continue; // should not happen + // Parse the selector + // Here is the relevant part of the CSS grammar: + // + // ruleset + // : selector [ ',' S* selector ]* '{' ... + // selector + // : simple_selector [ combinator selector | S+ [ combinator? selector ]? ]? + // combinator + // : '+' S* + // : '>' S* + // simple_selector + // : element_name [ HASH | class | attrib | pseudo ]* + // | [ HASH | class | attrib | pseudo ]+ + // element_name + // : IDENT | '*' + // ; + // class + // : '.' IDENT + // ; + // attrib + // : '[' S* IDENT S* [ [ '=' | INCLUDES | DASHMATCH ] S* + // [ IDENT | STRING ] S* ]? ']' + // ; + // pseudo + // : ':' [ IDENT | FUNCTION S* [IDENT S*]? ')' ] + // ; + // + // For reference, here are the relevant tokens: + // + // HASH #{name} + // IDENT {ident} + // INCLUDES == + // DASHMATCH |= + // STRING {string} + // FUNCTION {ident}\( + // + // And the lexical scanner tokens + // + // name {nmchar}+ + // nmchar [_a-z0-9-]|{nonascii}|{escape} + // nonascii [\240-\377] + // escape {unicode}|\\[^\r\n\f0-9a-f] + // unicode \\{h}}{1,6}(\r\n|[ \t\r\n\f])? + // ident -?{nmstart}{nmchar*} + // nmstart [_a-z]|{nonascii}|{escape} + // string {string1}|{string2} + // string1 \"([^\n\r\f\\"]|\\{nl}|{escape})*\" + // string2 \'([^\n\r\f\\"]|\\{nl}|{escape})*\' + // + // We'll implement a subset (in order to reduce attack + // surface); in particular: + // + // - No Unicode support + // - No escapes support + // - No string support (by proxy no attrib support) + // - element_name is matched against allowed + // elements (some people might find this + // annoying...) + // - Pseudo-elements one of :first-child, :link, + // :visited, :active, :hover, :focus + + // handle ruleset + $selectors = array_map('trim', explode(',', $selector)); + $new_selectors = array(); + foreach ($selectors as $sel) { + // split on +, > and spaces + $basic_selectors = preg_split('/\s*([+> ])\s*/', $sel, -1, PREG_SPLIT_DELIM_CAPTURE); + // even indices are chunks, odd indices are + // delimiters + $nsel = null; + $delim = null; // guaranteed to be non-null after + // two loop iterations + for ($i = 0, $c = count($basic_selectors); $i < $c; $i++) { + $x = $basic_selectors[$i]; + if ($i % 2) { + // delimiter + if ($x === ' ') { + $delim = ' '; + } else { + $delim = ' ' . $x . ' '; + } + } else { + // simple selector + $components = preg_split('/([#.:])/', $x, -1, PREG_SPLIT_DELIM_CAPTURE); + $sdelim = null; + $nx = null; + for ($j = 0, $cc = count($components); $j < $cc; $j ++) { + $y = $components[$j]; + if ($j === 0) { + if ($y === '*' || isset($html_definition->info[$y = strtolower($y)])) { + $nx = $y; + } else { + // $nx stays null; this matters + // if we don't manage to find + // any valid selector content, + // in which case we ignore the + // outer $delim + } + } elseif ($j % 2) { + // set delimiter + $sdelim = $y; + } else { + $attrdef = null; + if ($sdelim === '#') { + $attrdef = $this->_id_attrdef; + } elseif ($sdelim === '.') { + $attrdef = $this->_class_attrdef; + } elseif ($sdelim === ':') { + $attrdef = $this->_enum_attrdef; + } else { + throw new HTMLPurifier_Exception('broken invariant sdelim and preg_split'); + } + $r = $attrdef->validate($y, $config, $context); + if ($r !== false) { + if ($r !== true) { + $y = $r; + } + if ($nx === null) { + $nx = ''; + } + $nx .= $sdelim . $y; + } + } + } + if ($nx !== null) { + if ($nsel === null) { + $nsel = $nx; + } else { + $nsel .= $delim . $nx; + } + } else { + // delimiters to the left of invalid + // basic selector ignored + } + } + } + if ($nsel !== null) { + if (!empty($scopes)) { + foreach ($scopes as $s) { + $new_selectors[] = "$s $nsel"; + } + } else { + $new_selectors[] = $nsel; + } + } + } + if (empty($new_selectors)) continue; + $selector = implode(', ', $new_selectors); + foreach ($style as $name => $value) { + if (!isset($css_definition->info[$name])) { + unset($style[$name]); + continue; + } + $def = $css_definition->info[$name]; + $ret = $def->validate($value, $config, $context); + if ($ret === false) unset($style[$name]); + else $style[$name] = $ret; + } + $new_decls[$selector] = $style; + } + $new_css[$k] = $new_decls; + } + // remove stuff that shouldn't be used, could be reenabled + // after security risks are analyzed + $this->_tidy->css = $new_css; + $this->_tidy->import = array(); + $this->_tidy->charset = null; + $this->_tidy->namespace = null; + $css = $this->_tidy->print->plain(); + // we are going to escape any special characters <>& to ensure + // that no funny business occurs (i.e. </style> in a font-family prop). + if ($config->get('Filter.ExtractStyleBlocks.Escaping')) { + $css = str_replace( + array('<', '>', '&'), + array('\3C ', '\3E ', '\26 '), + $css + ); + } + return $css; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Filter/YouTube.php b/lib/htmlpurifier/library/HTMLPurifier/Filter/YouTube.php new file mode 100644 index 000000000..23df221ea --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Filter/YouTube.php @@ -0,0 +1,39 @@ +<?php + +class HTMLPurifier_Filter_YouTube extends HTMLPurifier_Filter +{ + + public $name = 'YouTube'; + + public function preFilter($html, $config, $context) { + $pre_regex = '#<object[^>]+>.+?'. + 'http://www.youtube.com/((?:v|cp)/[A-Za-z0-9\-_=]+).+?</object>#s'; + $pre_replace = '<span class="youtube-embed">\1</span>'; + return preg_replace($pre_regex, $pre_replace, $html); + } + + public function postFilter($html, $config, $context) { + $post_regex = '#<span class="youtube-embed">((?:v|cp)/[A-Za-z0-9\-_=]+)</span>#'; + return preg_replace_callback($post_regex, array($this, 'postFilterCallback'), $html); + } + + protected function armorUrl($url) { + return str_replace('--', '--', $url); + } + + protected function postFilterCallback($matches) { + $url = $this->armorUrl($matches[1]); + return '<object width="425" height="350" type="application/x-shockwave-flash" '. + 'data="http://www.youtube.com/'.$url.'">'. + '<param name="movie" value="http://www.youtube.com/'.$url.'"></param>'. + '<!--[if IE]>'. + '<embed src="http://www.youtube.com/'.$url.'"'. + 'type="application/x-shockwave-flash"'. + 'wmode="transparent" width="425" height="350" />'. + '<![endif]-->'. + '</object>'; + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Generator.php b/lib/htmlpurifier/library/HTMLPurifier/Generator.php new file mode 100644 index 000000000..fee1a5f84 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Generator.php @@ -0,0 +1,254 @@ +<?php + +/** + * Generates HTML from tokens. + * @todo Refactor interface so that configuration/context is determined + * upon instantiation, no need for messy generateFromTokens() calls + * @todo Make some of the more internal functions protected, and have + * unit tests work around that + */ +class HTMLPurifier_Generator +{ + + /** + * Whether or not generator should produce XML output + */ + private $_xhtml = true; + + /** + * :HACK: Whether or not generator should comment the insides of <script> tags + */ + private $_scriptFix = false; + + /** + * Cache of HTMLDefinition during HTML output to determine whether or + * not attributes should be minimized. + */ + private $_def; + + /** + * Cache of %Output.SortAttr + */ + private $_sortAttr; + + /** + * Cache of %Output.FlashCompat + */ + private $_flashCompat; + + /** + * Cache of %Output.FixInnerHTML + */ + private $_innerHTMLFix; + + /** + * Stack for keeping track of object information when outputting IE + * compatibility code. + */ + private $_flashStack = array(); + + /** + * Configuration for the generator + */ + protected $config; + + /** + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + */ + public function __construct($config, $context) { + $this->config = $config; + $this->_scriptFix = $config->get('Output.CommentScriptContents'); + $this->_innerHTMLFix = $config->get('Output.FixInnerHTML'); + $this->_sortAttr = $config->get('Output.SortAttr'); + $this->_flashCompat = $config->get('Output.FlashCompat'); + $this->_def = $config->getHTMLDefinition(); + $this->_xhtml = $this->_def->doctype->xml; + } + + /** + * Generates HTML from an array of tokens. + * @param $tokens Array of HTMLPurifier_Token + * @param $config HTMLPurifier_Config object + * @return Generated HTML + */ + public function generateFromTokens($tokens) { + if (!$tokens) return ''; + + // Basic algorithm + $html = ''; + for ($i = 0, $size = count($tokens); $i < $size; $i++) { + if ($this->_scriptFix && $tokens[$i]->name === 'script' + && $i + 2 < $size && $tokens[$i+2] instanceof HTMLPurifier_Token_End) { + // script special case + // the contents of the script block must be ONE token + // for this to work. + $html .= $this->generateFromToken($tokens[$i++]); + $html .= $this->generateScriptFromToken($tokens[$i++]); + } + $html .= $this->generateFromToken($tokens[$i]); + } + + // Tidy cleanup + if (extension_loaded('tidy') && $this->config->get('Output.TidyFormat')) { + $tidy = new Tidy; + $tidy->parseString($html, array( + 'indent'=> true, + 'output-xhtml' => $this->_xhtml, + 'show-body-only' => true, + 'indent-spaces' => 2, + 'wrap' => 68, + ), 'utf8'); + $tidy->cleanRepair(); + $html = (string) $tidy; // explicit cast necessary + } + + // Normalize newlines to system defined value + if ($this->config->get('Core.NormalizeNewlines')) { + $nl = $this->config->get('Output.Newline'); + if ($nl === null) $nl = PHP_EOL; + if ($nl !== "\n") $html = str_replace("\n", $nl, $html); + } + return $html; + } + + /** + * Generates HTML from a single token. + * @param $token HTMLPurifier_Token object. + * @return Generated HTML + */ + public function generateFromToken($token) { + if (!$token instanceof HTMLPurifier_Token) { + trigger_error('Cannot generate HTML from non-HTMLPurifier_Token object', E_USER_WARNING); + return ''; + + } elseif ($token instanceof HTMLPurifier_Token_Start) { + $attr = $this->generateAttributes($token->attr, $token->name); + if ($this->_flashCompat) { + if ($token->name == "object") { + $flash = new stdclass(); + $flash->attr = $token->attr; + $flash->param = array(); + $this->_flashStack[] = $flash; + } + } + return '<' . $token->name . ($attr ? ' ' : '') . $attr . '>'; + + } elseif ($token instanceof HTMLPurifier_Token_End) { + $_extra = ''; + if ($this->_flashCompat) { + if ($token->name == "object" && !empty($this->_flashStack)) { + // doesn't do anything for now + } + } + return $_extra . '</' . $token->name . '>'; + + } elseif ($token instanceof HTMLPurifier_Token_Empty) { + if ($this->_flashCompat && $token->name == "param" && !empty($this->_flashStack)) { + $this->_flashStack[count($this->_flashStack)-1]->param[$token->attr['name']] = $token->attr['value']; + } + $attr = $this->generateAttributes($token->attr, $token->name); + return '<' . $token->name . ($attr ? ' ' : '') . $attr . + ( $this->_xhtml ? ' /': '' ) // <br /> v. <br> + . '>'; + + } elseif ($token instanceof HTMLPurifier_Token_Text) { + return $this->escape($token->data, ENT_NOQUOTES); + + } elseif ($token instanceof HTMLPurifier_Token_Comment) { + return '<!--' . $token->data . '-->'; + } else { + return ''; + + } + } + + /** + * Special case processor for the contents of script tags + * @warning This runs into problems if there's already a literal + * --> somewhere inside the script contents. + */ + public function generateScriptFromToken($token) { + if (!$token instanceof HTMLPurifier_Token_Text) return $this->generateFromToken($token); + // Thanks <http://lachy.id.au/log/2005/05/script-comments> + $data = preg_replace('#//\s*$#', '', $token->data); + return '<!--//--><![CDATA[//><!--' . "\n" . trim($data) . "\n" . '//--><!]]>'; + } + + /** + * Generates attribute declarations from attribute array. + * @note This does not include the leading or trailing space. + * @param $assoc_array_of_attributes Attribute array + * @param $element Name of element attributes are for, used to check + * attribute minimization. + * @return Generate HTML fragment for insertion. + */ + public function generateAttributes($assoc_array_of_attributes, $element = false) { + $html = ''; + if ($this->_sortAttr) ksort($assoc_array_of_attributes); + foreach ($assoc_array_of_attributes as $key => $value) { + if (!$this->_xhtml) { + // Remove namespaced attributes + if (strpos($key, ':') !== false) continue; + // Check if we should minimize the attribute: val="val" -> val + if ($element && !empty($this->_def->info[$element]->attr[$key]->minimized)) { + $html .= $key . ' '; + continue; + } + } + // Workaround for Internet Explorer innerHTML bug. + // Essentially, Internet Explorer, when calculating + // innerHTML, omits quotes if there are no instances of + // angled brackets, quotes or spaces. However, when parsing + // HTML (for example, when you assign to innerHTML), it + // treats backticks as quotes. Thus, + // <img alt="``" /> + // becomes + // <img alt=`` /> + // becomes + // <img alt='' /> + // Fortunately, all we need to do is trigger an appropriate + // quoting style, which we do by adding an extra space. + // This also is consistent with the W3C spec, which states + // that user agents may ignore leading or trailing + // whitespace (in fact, most don't, at least for attributes + // like alt, but an extra space at the end is barely + // noticeable). Still, we have a configuration knob for + // this, since this transformation is not necesary if you + // don't process user input with innerHTML or you don't plan + // on supporting Internet Explorer. + if ($this->_innerHTMLFix) { + if (strpos($value, '`') !== false) { + // check if correct quoting style would not already be + // triggered + if (strcspn($value, '"\' <>') === strlen($value)) { + // protect! + $value .= ' '; + } + } + } + $html .= $key.'="'.$this->escape($value).'" '; + } + return rtrim($html); + } + + /** + * Escapes raw text data. + * @todo This really ought to be protected, but until we have a facility + * for properly generating HTML here w/o using tokens, it stays + * public. + * @param $string String data to escape for HTML. + * @param $quote Quoting style, like htmlspecialchars. ENT_NOQUOTES is + * permissible for non-attribute output. + * @return String escaped data. + */ + public function escape($string, $quote = null) { + // Workaround for APC bug on Mac Leopard reported by sidepodcast + // http://htmlpurifier.org/phorum/read.php?3,4823,4846 + if ($quote === null) $quote = ENT_COMPAT; + return htmlspecialchars($string, $quote, 'UTF-8'); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLDefinition.php new file mode 100644 index 000000000..b079d44c1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLDefinition.php @@ -0,0 +1,425 @@ +<?php + +/** + * Definition of the purified HTML that describes allowed children, + * attributes, and many other things. + * + * Conventions: + * + * All member variables that are prefixed with info + * (including the main $info array) are used by HTML Purifier internals + * and should not be directly edited when customizing the HTMLDefinition. + * They can usually be set via configuration directives or custom + * modules. + * + * On the other hand, member variables without the info prefix are used + * internally by the HTMLDefinition and MUST NOT be used by other HTML + * Purifier internals. Many of them, however, are public, and may be + * edited by userspace code to tweak the behavior of HTMLDefinition. + * + * @note This class is inspected by Printer_HTMLDefinition; please + * update that class if things here change. + * + * @warning Directives that change this object's structure must be in + * the HTML or Attr namespace! + */ +class HTMLPurifier_HTMLDefinition extends HTMLPurifier_Definition +{ + + // FULLY-PUBLIC VARIABLES --------------------------------------------- + + /** + * Associative array of element names to HTMLPurifier_ElementDef + */ + public $info = array(); + + /** + * Associative array of global attribute name to attribute definition. + */ + public $info_global_attr = array(); + + /** + * String name of parent element HTML will be going into. + */ + public $info_parent = 'div'; + + /** + * Definition for parent element, allows parent element to be a + * tag that's not allowed inside the HTML fragment. + */ + public $info_parent_def; + + /** + * String name of element used to wrap inline elements in block context + * @note This is rarely used except for BLOCKQUOTEs in strict mode + */ + public $info_block_wrapper = 'p'; + + /** + * Associative array of deprecated tag name to HTMLPurifier_TagTransform + */ + public $info_tag_transform = array(); + + /** + * Indexed list of HTMLPurifier_AttrTransform to be performed before validation. + */ + public $info_attr_transform_pre = array(); + + /** + * Indexed list of HTMLPurifier_AttrTransform to be performed after validation. + */ + public $info_attr_transform_post = array(); + + /** + * Nested lookup array of content set name (Block, Inline) to + * element name to whether or not it belongs in that content set. + */ + public $info_content_sets = array(); + + /** + * Indexed list of HTMLPurifier_Injector to be used. + */ + public $info_injector = array(); + + /** + * Doctype object + */ + public $doctype; + + + + // RAW CUSTOMIZATION STUFF -------------------------------------------- + + /** + * Adds a custom attribute to a pre-existing element + * @note This is strictly convenience, and does not have a corresponding + * method in HTMLPurifier_HTMLModule + * @param $element_name String element name to add attribute to + * @param $attr_name String name of attribute + * @param $def Attribute definition, can be string or object, see + * HTMLPurifier_AttrTypes for details + */ + public function addAttribute($element_name, $attr_name, $def) { + $module = $this->getAnonymousModule(); + if (!isset($module->info[$element_name])) { + $element = $module->addBlankElement($element_name); + } else { + $element = $module->info[$element_name]; + } + $element->attr[$attr_name] = $def; + } + + /** + * Adds a custom element to your HTML definition + * @note See HTMLPurifier_HTMLModule::addElement for detailed + * parameter and return value descriptions. + */ + public function addElement($element_name, $type, $contents, $attr_collections, $attributes = array()) { + $module = $this->getAnonymousModule(); + // assume that if the user is calling this, the element + // is safe. This may not be a good idea + $element = $module->addElement($element_name, $type, $contents, $attr_collections, $attributes); + return $element; + } + + /** + * Adds a blank element to your HTML definition, for overriding + * existing behavior + * @note See HTMLPurifier_HTMLModule::addBlankElement for detailed + * parameter and return value descriptions. + */ + public function addBlankElement($element_name) { + $module = $this->getAnonymousModule(); + $element = $module->addBlankElement($element_name); + return $element; + } + + /** + * Retrieves a reference to the anonymous module, so you can + * bust out advanced features without having to make your own + * module. + */ + public function getAnonymousModule() { + if (!$this->_anonModule) { + $this->_anonModule = new HTMLPurifier_HTMLModule(); + $this->_anonModule->name = 'Anonymous'; + } + return $this->_anonModule; + } + + private $_anonModule = null; + + + // PUBLIC BUT INTERNAL VARIABLES -------------------------------------- + + public $type = 'HTML'; + public $manager; /**< Instance of HTMLPurifier_HTMLModuleManager */ + + /** + * Performs low-cost, preliminary initialization. + */ + public function __construct() { + $this->manager = new HTMLPurifier_HTMLModuleManager(); + } + + protected function doSetup($config) { + $this->processModules($config); + $this->setupConfigStuff($config); + unset($this->manager); + + // cleanup some of the element definitions + foreach ($this->info as $k => $v) { + unset($this->info[$k]->content_model); + unset($this->info[$k]->content_model_type); + } + } + + /** + * Extract out the information from the manager + */ + protected function processModules($config) { + + if ($this->_anonModule) { + // for user specific changes + // this is late-loaded so we don't have to deal with PHP4 + // reference wonky-ness + $this->manager->addModule($this->_anonModule); + unset($this->_anonModule); + } + + $this->manager->setup($config); + $this->doctype = $this->manager->doctype; + + foreach ($this->manager->modules as $module) { + foreach($module->info_tag_transform as $k => $v) { + if ($v === false) unset($this->info_tag_transform[$k]); + else $this->info_tag_transform[$k] = $v; + } + foreach($module->info_attr_transform_pre as $k => $v) { + if ($v === false) unset($this->info_attr_transform_pre[$k]); + else $this->info_attr_transform_pre[$k] = $v; + } + foreach($module->info_attr_transform_post as $k => $v) { + if ($v === false) unset($this->info_attr_transform_post[$k]); + else $this->info_attr_transform_post[$k] = $v; + } + foreach ($module->info_injector as $k => $v) { + if ($v === false) unset($this->info_injector[$k]); + else $this->info_injector[$k] = $v; + } + } + + $this->info = $this->manager->getElements(); + $this->info_content_sets = $this->manager->contentSets->lookup; + + } + + /** + * Sets up stuff based on config. We need a better way of doing this. + */ + protected function setupConfigStuff($config) { + + $block_wrapper = $config->get('HTML.BlockWrapper'); + if (isset($this->info_content_sets['Block'][$block_wrapper])) { + $this->info_block_wrapper = $block_wrapper; + } else { + trigger_error('Cannot use non-block element as block wrapper', + E_USER_ERROR); + } + + $parent = $config->get('HTML.Parent'); + $def = $this->manager->getElement($parent, true); + if ($def) { + $this->info_parent = $parent; + $this->info_parent_def = $def; + } else { + trigger_error('Cannot use unrecognized element as parent', + E_USER_ERROR); + $this->info_parent_def = $this->manager->getElement($this->info_parent, true); + } + + // support template text + $support = "(for information on implementing this, see the ". + "support forums) "; + + // setup allowed elements ----------------------------------------- + + $allowed_elements = $config->get('HTML.AllowedElements'); + $allowed_attributes = $config->get('HTML.AllowedAttributes'); // retrieve early + + if (!is_array($allowed_elements) && !is_array($allowed_attributes)) { + $allowed = $config->get('HTML.Allowed'); + if (is_string($allowed)) { + list($allowed_elements, $allowed_attributes) = $this->parseTinyMCEAllowedList($allowed); + } + } + + if (is_array($allowed_elements)) { + foreach ($this->info as $name => $d) { + if(!isset($allowed_elements[$name])) unset($this->info[$name]); + unset($allowed_elements[$name]); + } + // emit errors + foreach ($allowed_elements as $element => $d) { + $element = htmlspecialchars($element); // PHP doesn't escape errors, be careful! + trigger_error("Element '$element' is not supported $support", E_USER_WARNING); + } + } + + // setup allowed attributes --------------------------------------- + + $allowed_attributes_mutable = $allowed_attributes; // by copy! + if (is_array($allowed_attributes)) { + + // This actually doesn't do anything, since we went away from + // global attributes. It's possible that userland code uses + // it, but HTMLModuleManager doesn't! + foreach ($this->info_global_attr as $attr => $x) { + $keys = array($attr, "*@$attr", "*.$attr"); + $delete = true; + foreach ($keys as $key) { + if ($delete && isset($allowed_attributes[$key])) { + $delete = false; + } + if (isset($allowed_attributes_mutable[$key])) { + unset($allowed_attributes_mutable[$key]); + } + } + if ($delete) unset($this->info_global_attr[$attr]); + } + + foreach ($this->info as $tag => $info) { + foreach ($info->attr as $attr => $x) { + $keys = array("$tag@$attr", $attr, "*@$attr", "$tag.$attr", "*.$attr"); + $delete = true; + foreach ($keys as $key) { + if ($delete && isset($allowed_attributes[$key])) { + $delete = false; + } + if (isset($allowed_attributes_mutable[$key])) { + unset($allowed_attributes_mutable[$key]); + } + } + if ($delete) { + if ($this->info[$tag]->attr[$attr]->required) { + trigger_error("Required attribute '$attr' in element '$tag' was not allowed, which means '$tag' will not be allowed either", E_USER_WARNING); + } + unset($this->info[$tag]->attr[$attr]); + } + } + } + // emit errors + foreach ($allowed_attributes_mutable as $elattr => $d) { + $bits = preg_split('/[.@]/', $elattr, 2); + $c = count($bits); + switch ($c) { + case 2: + if ($bits[0] !== '*') { + $element = htmlspecialchars($bits[0]); + $attribute = htmlspecialchars($bits[1]); + if (!isset($this->info[$element])) { + trigger_error("Cannot allow attribute '$attribute' if element '$element' is not allowed/supported $support"); + } else { + trigger_error("Attribute '$attribute' in element '$element' not supported $support", + E_USER_WARNING); + } + break; + } + // otherwise fall through + case 1: + $attribute = htmlspecialchars($bits[0]); + trigger_error("Global attribute '$attribute' is not ". + "supported in any elements $support", + E_USER_WARNING); + break; + } + } + + } + + // setup forbidden elements --------------------------------------- + + $forbidden_elements = $config->get('HTML.ForbiddenElements'); + $forbidden_attributes = $config->get('HTML.ForbiddenAttributes'); + + foreach ($this->info as $tag => $info) { + if (isset($forbidden_elements[$tag])) { + unset($this->info[$tag]); + continue; + } + foreach ($info->attr as $attr => $x) { + if ( + isset($forbidden_attributes["$tag@$attr"]) || + isset($forbidden_attributes["*@$attr"]) || + isset($forbidden_attributes[$attr]) + ) { + unset($this->info[$tag]->attr[$attr]); + continue; + } // this segment might get removed eventually + elseif (isset($forbidden_attributes["$tag.$attr"])) { + // $tag.$attr are not user supplied, so no worries! + trigger_error("Error with $tag.$attr: tag.attr syntax not supported for HTML.ForbiddenAttributes; use tag@attr instead", E_USER_WARNING); + } + } + } + foreach ($forbidden_attributes as $key => $v) { + if (strlen($key) < 2) continue; + if ($key[0] != '*') continue; + if ($key[1] == '.') { + trigger_error("Error with $key: *.attr syntax not supported for HTML.ForbiddenAttributes; use attr instead", E_USER_WARNING); + } + } + + // setup injectors ----------------------------------------------------- + foreach ($this->info_injector as $i => $injector) { + if ($injector->checkNeeded($config) !== false) { + // remove injector that does not have it's required + // elements/attributes present, and is thus not needed. + unset($this->info_injector[$i]); + } + } + } + + /** + * Parses a TinyMCE-flavored Allowed Elements and Attributes list into + * separate lists for processing. Format is element[attr1|attr2],element2... + * @warning Although it's largely drawn from TinyMCE's implementation, + * it is different, and you'll probably have to modify your lists + * @param $list String list to parse + * @param array($allowed_elements, $allowed_attributes) + * @todo Give this its own class, probably static interface + */ + public function parseTinyMCEAllowedList($list) { + + $list = str_replace(array(' ', "\t"), '', $list); + + $elements = array(); + $attributes = array(); + + $chunks = preg_split('/(,|[\n\r]+)/', $list); + foreach ($chunks as $chunk) { + if (empty($chunk)) continue; + // remove TinyMCE element control characters + if (!strpos($chunk, '[')) { + $element = $chunk; + $attr = false; + } else { + list($element, $attr) = explode('[', $chunk); + } + if ($element !== '*') $elements[$element] = true; + if (!$attr) continue; + $attr = substr($attr, 0, strlen($attr) - 1); // remove trailing ] + $attr = explode('|', $attr); + foreach ($attr as $key) { + $attributes["$element.$key"] = true; + } + } + + return array($elements, $attributes); + + } + + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule.php new file mode 100644 index 000000000..072cf6808 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule.php @@ -0,0 +1,244 @@ +<?php + +/** + * Represents an XHTML 1.1 module, with information on elements, tags + * and attributes. + * @note Even though this is technically XHTML 1.1, it is also used for + * regular HTML parsing. We are using modulization as a convenient + * way to represent the internals of HTMLDefinition, and our + * implementation is by no means conforming and does not directly + * use the normative DTDs or XML schemas. + * @note The public variables in a module should almost directly + * correspond to the variables in HTMLPurifier_HTMLDefinition. + * However, the prefix info carries no special meaning in these + * objects (include it anyway if that's the correspondence though). + * @todo Consider making some member functions protected + */ + +class HTMLPurifier_HTMLModule +{ + + // -- Overloadable ---------------------------------------------------- + + /** + * Short unique string identifier of the module + */ + public $name; + + /** + * Informally, a list of elements this module changes. Not used in + * any significant way. + */ + public $elements = array(); + + /** + * Associative array of element names to element definitions. + * Some definitions may be incomplete, to be merged in later + * with the full definition. + */ + public $info = array(); + + /** + * Associative array of content set names to content set additions. + * This is commonly used to, say, add an A element to the Inline + * content set. This corresponds to an internal variable $content_sets + * and NOT info_content_sets member variable of HTMLDefinition. + */ + public $content_sets = array(); + + /** + * Associative array of attribute collection names to attribute + * collection additions. More rarely used for adding attributes to + * the global collections. Example is the StyleAttribute module adding + * the style attribute to the Core. Corresponds to HTMLDefinition's + * attr_collections->info, since the object's data is only info, + * with extra behavior associated with it. + */ + public $attr_collections = array(); + + /** + * Associative array of deprecated tag name to HTMLPurifier_TagTransform + */ + public $info_tag_transform = array(); + + /** + * List of HTMLPurifier_AttrTransform to be performed before validation. + */ + public $info_attr_transform_pre = array(); + + /** + * List of HTMLPurifier_AttrTransform to be performed after validation. + */ + public $info_attr_transform_post = array(); + + /** + * List of HTMLPurifier_Injector to be performed during well-formedness fixing. + * An injector will only be invoked if all of it's pre-requisites are met; + * if an injector fails setup, there will be no error; it will simply be + * silently disabled. + */ + public $info_injector = array(); + + /** + * Boolean flag that indicates whether or not getChildDef is implemented. + * For optimization reasons: may save a call to a function. Be sure + * to set it if you do implement getChildDef(), otherwise it will have + * no effect! + */ + public $defines_child_def = false; + + /** + * Boolean flag whether or not this module is safe. If it is not safe, all + * of its members are unsafe. Modules are safe by default (this might be + * slightly dangerous, but it doesn't make much sense to force HTML Purifier, + * which is based off of safe HTML, to explicitly say, "This is safe," even + * though there are modules which are "unsafe") + * + * @note Previously, safety could be applied at an element level granularity. + * We've removed this ability, so in order to add "unsafe" elements + * or attributes, a dedicated module with this property set to false + * must be used. + */ + public $safe = true; + + /** + * Retrieves a proper HTMLPurifier_ChildDef subclass based on + * content_model and content_model_type member variables of + * the HTMLPurifier_ElementDef class. There is a similar function + * in HTMLPurifier_HTMLDefinition. + * @param $def HTMLPurifier_ElementDef instance + * @return HTMLPurifier_ChildDef subclass + */ + public function getChildDef($def) {return false;} + + // -- Convenience ----------------------------------------------------- + + /** + * Convenience function that sets up a new element + * @param $element Name of element to add + * @param $type What content set should element be registered to? + * Set as false to skip this step. + * @param $contents Allowed children in form of: + * "$content_model_type: $content_model" + * @param $attr_includes What attribute collections to register to + * element? + * @param $attr What unique attributes does the element define? + * @note See ElementDef for in-depth descriptions of these parameters. + * @return Created element definition object, so you + * can set advanced parameters + */ + public function addElement($element, $type, $contents, $attr_includes = array(), $attr = array()) { + $this->elements[] = $element; + // parse content_model + list($content_model_type, $content_model) = $this->parseContents($contents); + // merge in attribute inclusions + $this->mergeInAttrIncludes($attr, $attr_includes); + // add element to content sets + if ($type) $this->addElementToContentSet($element, $type); + // create element + $this->info[$element] = HTMLPurifier_ElementDef::create( + $content_model, $content_model_type, $attr + ); + // literal object $contents means direct child manipulation + if (!is_string($contents)) $this->info[$element]->child = $contents; + return $this->info[$element]; + } + + /** + * Convenience function that creates a totally blank, non-standalone + * element. + * @param $element Name of element to create + * @return Created element + */ + public function addBlankElement($element) { + if (!isset($this->info[$element])) { + $this->elements[] = $element; + $this->info[$element] = new HTMLPurifier_ElementDef(); + $this->info[$element]->standalone = false; + } else { + trigger_error("Definition for $element already exists in module, cannot redefine"); + } + return $this->info[$element]; + } + + /** + * Convenience function that registers an element to a content set + * @param Element to register + * @param Name content set (warning: case sensitive, usually upper-case + * first letter) + */ + public function addElementToContentSet($element, $type) { + if (!isset($this->content_sets[$type])) $this->content_sets[$type] = ''; + else $this->content_sets[$type] .= ' | '; + $this->content_sets[$type] .= $element; + } + + /** + * Convenience function that transforms single-string contents + * into separate content model and content model type + * @param $contents Allowed children in form of: + * "$content_model_type: $content_model" + * @note If contents is an object, an array of two nulls will be + * returned, and the callee needs to take the original $contents + * and use it directly. + */ + public function parseContents($contents) { + if (!is_string($contents)) return array(null, null); // defer + switch ($contents) { + // check for shorthand content model forms + case 'Empty': + return array('empty', ''); + case 'Inline': + return array('optional', 'Inline | #PCDATA'); + case 'Flow': + return array('optional', 'Flow | #PCDATA'); + } + list($content_model_type, $content_model) = explode(':', $contents); + $content_model_type = strtolower(trim($content_model_type)); + $content_model = trim($content_model); + return array($content_model_type, $content_model); + } + + /** + * Convenience function that merges a list of attribute includes into + * an attribute array. + * @param $attr Reference to attr array to modify + * @param $attr_includes Array of includes / string include to merge in + */ + public function mergeInAttrIncludes(&$attr, $attr_includes) { + if (!is_array($attr_includes)) { + if (empty($attr_includes)) $attr_includes = array(); + else $attr_includes = array($attr_includes); + } + $attr[0] = $attr_includes; + } + + /** + * Convenience function that generates a lookup table with boolean + * true as value. + * @param $list List of values to turn into a lookup + * @note You can also pass an arbitrary number of arguments in + * place of the regular argument + * @return Lookup array equivalent of list + */ + public function makeLookup($list) { + if (is_string($list)) $list = func_get_args(); + $ret = array(); + foreach ($list as $value) { + if (is_null($value)) continue; + $ret[$value] = true; + } + return $ret; + } + + /** + * Lazy load construction of the module after determining whether + * or not it's needed, and also when a finalized configuration object + * is available. + * @param $config Instance of HTMLPurifier_Config + */ + public function setup($config) {} + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Bdo.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Bdo.php new file mode 100644 index 000000000..3d66f1b4e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Bdo.php @@ -0,0 +1,31 @@ +<?php + +/** + * XHTML 1.1 Bi-directional Text Module, defines elements that + * declare directionality of content. Text Extension Module. + */ +class HTMLPurifier_HTMLModule_Bdo extends HTMLPurifier_HTMLModule +{ + + public $name = 'Bdo'; + public $attr_collections = array( + 'I18N' => array('dir' => false) + ); + + public function setup($config) { + $bdo = $this->addElement( + 'bdo', 'Inline', 'Inline', array('Core', 'Lang'), + array( + 'dir' => 'Enum#ltr,rtl', // required + // The Abstract Module specification has the attribute + // inclusions wrong for bdo: bdo allows Lang + ) + ); + $bdo->attr_transform_post['required-dir'] = new HTMLPurifier_AttrTransform_BdoDir(); + + $this->attr_collections['I18N']['dir'] = 'Enum#ltr,rtl'; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/CommonAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/CommonAttributes.php new file mode 100644 index 000000000..7c15da84f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/CommonAttributes.php @@ -0,0 +1,26 @@ +<?php + +class HTMLPurifier_HTMLModule_CommonAttributes extends HTMLPurifier_HTMLModule +{ + public $name = 'CommonAttributes'; + + public $attr_collections = array( + 'Core' => array( + 0 => array('Style'), + // 'xml:space' => false, + 'class' => 'Class', + 'id' => 'ID', + 'title' => 'CDATA', + ), + 'Lang' => array(), + 'I18N' => array( + 0 => array('Lang'), // proprietary, for xml:lang/lang + ), + 'Common' => array( + 0 => array('Core', 'I18N') + ) + ); + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Edit.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Edit.php new file mode 100644 index 000000000..ff9369055 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Edit.php @@ -0,0 +1,38 @@ +<?php + +/** + * XHTML 1.1 Edit Module, defines editing-related elements. Text Extension + * Module. + */ +class HTMLPurifier_HTMLModule_Edit extends HTMLPurifier_HTMLModule +{ + + public $name = 'Edit'; + + public function setup($config) { + $contents = 'Chameleon: #PCDATA | Inline ! #PCDATA | Flow'; + $attr = array( + 'cite' => 'URI', + // 'datetime' => 'Datetime', // not implemented + ); + $this->addElement('del', 'Inline', $contents, 'Common', $attr); + $this->addElement('ins', 'Inline', $contents, 'Common', $attr); + } + + // HTML 4.01 specifies that ins/del must not contain block + // elements when used in an inline context, chameleon is + // a complicated workaround to acheive this effect + + // Inline context ! Block context (exclamation mark is + // separator, see getChildDef for parsing) + + public $defines_child_def = true; + public function getChildDef($def) { + if ($def->content_model_type != 'chameleon') return false; + $value = explode('!', $def->content_model); + return new HTMLPurifier_ChildDef_Chameleon($value[0], $value[1]); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Forms.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Forms.php new file mode 100644 index 000000000..b963529a7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Forms.php @@ -0,0 +1,119 @@ +<?php + +/** + * XHTML 1.1 Forms module, defines all form-related elements found in HTML 4. + */ +class HTMLPurifier_HTMLModule_Forms extends HTMLPurifier_HTMLModule +{ + public $name = 'Forms'; + public $safe = false; + + public $content_sets = array( + 'Block' => 'Form', + 'Inline' => 'Formctrl', + ); + + public function setup($config) { + $form = $this->addElement('form', 'Form', + 'Required: Heading | List | Block | fieldset', 'Common', array( + 'accept' => 'ContentTypes', + 'accept-charset' => 'Charsets', + 'action*' => 'URI', + 'method' => 'Enum#get,post', + // really ContentType, but these two are the only ones used today + 'enctype' => 'Enum#application/x-www-form-urlencoded,multipart/form-data', + )); + $form->excludes = array('form' => true); + + $input = $this->addElement('input', 'Formctrl', 'Empty', 'Common', array( + 'accept' => 'ContentTypes', + 'accesskey' => 'Character', + 'alt' => 'Text', + 'checked' => 'Bool#checked', + 'disabled' => 'Bool#disabled', + 'maxlength' => 'Number', + 'name' => 'CDATA', + 'readonly' => 'Bool#readonly', + 'size' => 'Number', + 'src' => 'URI#embedded', + 'tabindex' => 'Number', + 'type' => 'Enum#text,password,checkbox,button,radio,submit,reset,file,hidden,image', + 'value' => 'CDATA', + )); + $input->attr_transform_post[] = new HTMLPurifier_AttrTransform_Input(); + + $this->addElement('select', 'Formctrl', 'Required: optgroup | option', 'Common', array( + 'disabled' => 'Bool#disabled', + 'multiple' => 'Bool#multiple', + 'name' => 'CDATA', + 'size' => 'Number', + 'tabindex' => 'Number', + )); + + $this->addElement('option', false, 'Optional: #PCDATA', 'Common', array( + 'disabled' => 'Bool#disabled', + 'label' => 'Text', + 'selected' => 'Bool#selected', + 'value' => 'CDATA', + )); + // It's illegal for there to be more than one selected, but not + // be multiple. Also, no selected means undefined behavior. This might + // be difficult to implement; perhaps an injector, or a context variable. + + $textarea = $this->addElement('textarea', 'Formctrl', 'Optional: #PCDATA', 'Common', array( + 'accesskey' => 'Character', + 'cols*' => 'Number', + 'disabled' => 'Bool#disabled', + 'name' => 'CDATA', + 'readonly' => 'Bool#readonly', + 'rows*' => 'Number', + 'tabindex' => 'Number', + )); + $textarea->attr_transform_pre[] = new HTMLPurifier_AttrTransform_Textarea(); + + $button = $this->addElement('button', 'Formctrl', 'Optional: #PCDATA | Heading | List | Block | Inline', 'Common', array( + 'accesskey' => 'Character', + 'disabled' => 'Bool#disabled', + 'name' => 'CDATA', + 'tabindex' => 'Number', + 'type' => 'Enum#button,submit,reset', + 'value' => 'CDATA', + )); + + // For exclusions, ideally we'd specify content sets, not literal elements + $button->excludes = $this->makeLookup( + 'form', 'fieldset', // Form + 'input', 'select', 'textarea', 'label', 'button', // Formctrl + 'a', // as per HTML 4.01 spec, this is omitted by modularization + 'isindex', 'iframe' // legacy items + ); + + // Extra exclusion: img usemap="" is not permitted within this element. + // We'll omit this for now, since we don't have any good way of + // indicating it yet. + + // This is HIGHLY user-unfriendly; we need a custom child-def for this + $this->addElement('fieldset', 'Form', 'Custom: (#WS?,legend,(Flow|#PCDATA)*)', 'Common'); + + $label = $this->addElement('label', 'Formctrl', 'Optional: #PCDATA | Inline', 'Common', array( + 'accesskey' => 'Character', + // 'for' => 'IDREF', // IDREF not implemented, cannot allow + )); + $label->excludes = array('label' => true); + + $this->addElement('legend', false, 'Optional: #PCDATA | Inline', 'Common', array( + 'accesskey' => 'Character', + )); + + $this->addElement('optgroup', false, 'Required: option', 'Common', array( + 'disabled' => 'Bool#disabled', + 'label*' => 'Text', + )); + + // Don't forget an injector for <isindex>. This one's a little complex + // because it maps to multiple elements. + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Hypertext.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Hypertext.php new file mode 100644 index 000000000..d7e9bdd27 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Hypertext.php @@ -0,0 +1,31 @@ +<?php + +/** + * XHTML 1.1 Hypertext Module, defines hypertext links. Core Module. + */ +class HTMLPurifier_HTMLModule_Hypertext extends HTMLPurifier_HTMLModule +{ + + public $name = 'Hypertext'; + + public function setup($config) { + $a = $this->addElement( + 'a', 'Inline', 'Inline', 'Common', + array( + // 'accesskey' => 'Character', + // 'charset' => 'Charset', + 'href' => 'URI', + // 'hreflang' => 'LanguageCode', + 'rel' => new HTMLPurifier_AttrDef_HTML_LinkTypes('rel'), + 'rev' => new HTMLPurifier_AttrDef_HTML_LinkTypes('rev'), + // 'tabindex' => 'Number', + // 'type' => 'ContentType', + ) + ); + $a->formatting = true; + $a->excludes = array('a' => true); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Iframe.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Iframe.php new file mode 100644 index 000000000..287071edf --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Iframe.php @@ -0,0 +1,38 @@ +<?php + +/** + * XHTML 1.1 Iframe Module provides inline frames. + * + * @note This module is not considered safe unless an Iframe + * whitelisting mechanism is specified. Currently, the only + * such mechanism is %URL.SafeIframeRegexp + */ +class HTMLPurifier_HTMLModule_Iframe extends HTMLPurifier_HTMLModule +{ + + public $name = 'Iframe'; + public $safe = false; + + public function setup($config) { + if ($config->get('HTML.SafeIframe')) { + $this->safe = true; + } + $this->addElement( + 'iframe', 'Inline', 'Flow', 'Common', + array( + 'src' => 'URI#embedded', + 'width' => 'Length', + 'height' => 'Length', + 'name' => 'ID', + 'scrolling' => 'Enum#yes,no,auto', + 'frameborder' => 'Enum#0,1', + 'longdesc' => 'URI', + 'marginheight' => 'Pixels', + 'marginwidth' => 'Pixels', + ) + ); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Image.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Image.php new file mode 100644 index 000000000..948d435bc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Image.php @@ -0,0 +1,40 @@ +<?php + +/** + * XHTML 1.1 Image Module provides basic image embedding. + * @note There is specialized code for removing empty images in + * HTMLPurifier_Strategy_RemoveForeignElements + */ +class HTMLPurifier_HTMLModule_Image extends HTMLPurifier_HTMLModule +{ + + public $name = 'Image'; + + public function setup($config) { + $max = $config->get('HTML.MaxImgLength'); + $img = $this->addElement( + 'img', 'Inline', 'Empty', 'Common', + array( + 'alt*' => 'Text', + // According to the spec, it's Length, but percents can + // be abused, so we allow only Pixels. + 'height' => 'Pixels#' . $max, + 'width' => 'Pixels#' . $max, + 'longdesc' => 'URI', + 'src*' => new HTMLPurifier_AttrDef_URI(true), // embedded + ) + ); + if ($max === null || $config->get('HTML.Trusted')) { + $img->attr['height'] = + $img->attr['width'] = 'Length'; + } + + // kind of strange, but splitting things up would be inefficient + $img->attr_transform_pre[] = + $img->attr_transform_post[] = + new HTMLPurifier_AttrTransform_ImgRequired(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Legacy.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Legacy.php new file mode 100644 index 000000000..f278eeced --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Legacy.php @@ -0,0 +1,159 @@ +<?php + +/** + * XHTML 1.1 Legacy module defines elements that were previously + * deprecated. + * + * @note Not all legacy elements have been implemented yet, which + * is a bit of a reverse problem as compared to browsers! In + * addition, this legacy module may implement a bit more than + * mandated by XHTML 1.1. + * + * This module can be used in combination with TransformToStrict in order + * to transform as many deprecated elements as possible, but retain + * questionably deprecated elements that do not have good alternatives + * as well as transform elements that don't have an implementation. + * See docs/ref-strictness.txt for more details. + */ + +class HTMLPurifier_HTMLModule_Legacy extends HTMLPurifier_HTMLModule +{ + + public $name = 'Legacy'; + + public function setup($config) { + + $this->addElement('basefont', 'Inline', 'Empty', false, array( + 'color' => 'Color', + 'face' => 'Text', // extremely broad, we should + 'size' => 'Text', // tighten it + 'id' => 'ID' + )); + $this->addElement('center', 'Block', 'Flow', 'Common'); + $this->addElement('dir', 'Block', 'Required: li', 'Common', array( + 'compact' => 'Bool#compact' + )); + $this->addElement('font', 'Inline', 'Inline', array('Core', 'I18N'), array( + 'color' => 'Color', + 'face' => 'Text', // extremely broad, we should + 'size' => 'Text', // tighten it + )); + $this->addElement('menu', 'Block', 'Required: li', 'Common', array( + 'compact' => 'Bool#compact' + )); + + $s = $this->addElement('s', 'Inline', 'Inline', 'Common'); + $s->formatting = true; + + $strike = $this->addElement('strike', 'Inline', 'Inline', 'Common'); + $strike->formatting = true; + + $u = $this->addElement('u', 'Inline', 'Inline', 'Common'); + $u->formatting = true; + + // setup modifications to old elements + + $align = 'Enum#left,right,center,justify'; + + $address = $this->addBlankElement('address'); + $address->content_model = 'Inline | #PCDATA | p'; + $address->content_model_type = 'optional'; + $address->child = false; + + $blockquote = $this->addBlankElement('blockquote'); + $blockquote->content_model = 'Flow | #PCDATA'; + $blockquote->content_model_type = 'optional'; + $blockquote->child = false; + + $br = $this->addBlankElement('br'); + $br->attr['clear'] = 'Enum#left,all,right,none'; + + $caption = $this->addBlankElement('caption'); + $caption->attr['align'] = 'Enum#top,bottom,left,right'; + + $div = $this->addBlankElement('div'); + $div->attr['align'] = $align; + + $dl = $this->addBlankElement('dl'); + $dl->attr['compact'] = 'Bool#compact'; + + for ($i = 1; $i <= 6; $i++) { + $h = $this->addBlankElement("h$i"); + $h->attr['align'] = $align; + } + + $hr = $this->addBlankElement('hr'); + $hr->attr['align'] = $align; + $hr->attr['noshade'] = 'Bool#noshade'; + $hr->attr['size'] = 'Pixels'; + $hr->attr['width'] = 'Length'; + + $img = $this->addBlankElement('img'); + $img->attr['align'] = 'IAlign'; + $img->attr['border'] = 'Pixels'; + $img->attr['hspace'] = 'Pixels'; + $img->attr['vspace'] = 'Pixels'; + + // figure out this integer business + + $li = $this->addBlankElement('li'); + $li->attr['value'] = new HTMLPurifier_AttrDef_Integer(); + $li->attr['type'] = 'Enum#s:1,i,I,a,A,disc,square,circle'; + + $ol = $this->addBlankElement('ol'); + $ol->attr['compact'] = 'Bool#compact'; + $ol->attr['start'] = new HTMLPurifier_AttrDef_Integer(); + $ol->attr['type'] = 'Enum#s:1,i,I,a,A'; + + $p = $this->addBlankElement('p'); + $p->attr['align'] = $align; + + $pre = $this->addBlankElement('pre'); + $pre->attr['width'] = 'Number'; + + // script omitted + + $table = $this->addBlankElement('table'); + $table->attr['align'] = 'Enum#left,center,right'; + $table->attr['bgcolor'] = 'Color'; + + $tr = $this->addBlankElement('tr'); + $tr->attr['bgcolor'] = 'Color'; + + $th = $this->addBlankElement('th'); + $th->attr['bgcolor'] = 'Color'; + $th->attr['height'] = 'Length'; + $th->attr['nowrap'] = 'Bool#nowrap'; + $th->attr['width'] = 'Length'; + + $td = $this->addBlankElement('td'); + $td->attr['bgcolor'] = 'Color'; + $td->attr['height'] = 'Length'; + $td->attr['nowrap'] = 'Bool#nowrap'; + $td->attr['width'] = 'Length'; + + $ul = $this->addBlankElement('ul'); + $ul->attr['compact'] = 'Bool#compact'; + $ul->attr['type'] = 'Enum#square,disc,circle'; + + // "safe" modifications to "unsafe" elements + // WARNING: If you want to add support for an unsafe, legacy + // attribute, make a new TrustedLegacy module with the trusted + // bit set appropriately + + $form = $this->addBlankElement('form'); + $form->content_model = 'Flow | #PCDATA'; + $form->content_model_type = 'optional'; + $form->attr['target'] = 'FrameTarget'; + + $input = $this->addBlankElement('input'); + $input->attr['align'] = 'IAlign'; + + $legend = $this->addBlankElement('legend'); + $legend->attr['align'] = 'LAlign'; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/List.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/List.php new file mode 100644 index 000000000..79ccefafd --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/List.php @@ -0,0 +1,43 @@ +<?php + +/** + * XHTML 1.1 List Module, defines list-oriented elements. Core Module. + */ +class HTMLPurifier_HTMLModule_List extends HTMLPurifier_HTMLModule +{ + + public $name = 'List'; + + // According to the abstract schema, the List content set is a fully formed + // one or more expr, but it invariably occurs in an optional declaration + // so we're not going to do that subtlety. It might cause trouble + // if a user defines "List" and expects that multiple lists are + // allowed to be specified, but then again, that's not very intuitive. + // Furthermore, the actual XML Schema may disagree. Regardless, + // we don't have support for such nested expressions without using + // the incredibly inefficient and draconic Custom ChildDef. + + public $content_sets = array('Flow' => 'List'); + + public function setup($config) { + $ol = $this->addElement('ol', 'List', new HTMLPurifier_ChildDef_List(), 'Common'); + $ul = $this->addElement('ul', 'List', new HTMLPurifier_ChildDef_List(), 'Common'); + // XXX The wrap attribute is handled by MakeWellFormed. This is all + // quite unsatisfactory, because we generated this + // *specifically* for lists, and now a big chunk of the handling + // is done properly by the List ChildDef. So actually, we just + // want enough information to make autoclosing work properly, + // and then hand off the tricky stuff to the ChildDef. + $ol->wrap = 'li'; + $ul->wrap = 'li'; + $this->addElement('dl', 'List', 'Required: dt | dd', 'Common'); + + $this->addElement('li', false, 'Flow', 'Common'); + + $this->addElement('dd', false, 'Flow', 'Common'); + $this->addElement('dt', false, 'Inline', 'Common'); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Name.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Name.php new file mode 100644 index 000000000..05694b450 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Name.php @@ -0,0 +1,21 @@ +<?php + +class HTMLPurifier_HTMLModule_Name extends HTMLPurifier_HTMLModule +{ + + public $name = 'Name'; + + public function setup($config) { + $elements = array('a', 'applet', 'form', 'frame', 'iframe', 'img', 'map'); + foreach ($elements as $name) { + $element = $this->addBlankElement($name); + $element->attr['name'] = 'CDATA'; + if (!$config->get('HTML.Attr.Name.UseCDATA')) { + $element->attr_transform_post['NameSync'] = new HTMLPurifier_AttrTransform_NameSync(); + } + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Nofollow.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Nofollow.php new file mode 100644 index 000000000..3aa6654a5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Nofollow.php @@ -0,0 +1,19 @@ +<?php + +/** + * Module adds the nofollow attribute transformation to a tags. It + * is enabled by HTML.Nofollow + */ +class HTMLPurifier_HTMLModule_Nofollow extends HTMLPurifier_HTMLModule +{ + + public $name = 'Nofollow'; + + public function setup($config) { + $a = $this->addBlankElement('a'); + $a->attr_transform_post[] = new HTMLPurifier_AttrTransform_Nofollow(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php new file mode 100644 index 000000000..5f1b14abb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/NonXMLCommonAttributes.php @@ -0,0 +1,14 @@ +<?php + +class HTMLPurifier_HTMLModule_NonXMLCommonAttributes extends HTMLPurifier_HTMLModule +{ + public $name = 'NonXMLCommonAttributes'; + + public $attr_collections = array( + 'Lang' => array( + 'lang' => 'LanguageCode', + ) + ); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Object.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Object.php new file mode 100644 index 000000000..193c1011f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Object.php @@ -0,0 +1,47 @@ +<?php + +/** + * XHTML 1.1 Object Module, defines elements for generic object inclusion + * @warning Users will commonly use <embed> to cater to legacy browsers: this + * module does not allow this sort of behavior + */ +class HTMLPurifier_HTMLModule_Object extends HTMLPurifier_HTMLModule +{ + + public $name = 'Object'; + public $safe = false; + + public function setup($config) { + + $this->addElement('object', 'Inline', 'Optional: #PCDATA | Flow | param', 'Common', + array( + 'archive' => 'URI', + 'classid' => 'URI', + 'codebase' => 'URI', + 'codetype' => 'Text', + 'data' => 'URI', + 'declare' => 'Bool#declare', + 'height' => 'Length', + 'name' => 'CDATA', + 'standby' => 'Text', + 'tabindex' => 'Number', + 'type' => 'ContentType', + 'width' => 'Length' + ) + ); + + $this->addElement('param', false, 'Empty', false, + array( + 'id' => 'ID', + 'name*' => 'Text', + 'type' => 'Text', + 'value' => 'Text', + 'valuetype' => 'Enum#data,ref,object' + ) + ); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Presentation.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Presentation.php new file mode 100644 index 000000000..8ff0b5ed7 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Presentation.php @@ -0,0 +1,36 @@ +<?php + +/** + * XHTML 1.1 Presentation Module, defines simple presentation-related + * markup. Text Extension Module. + * @note The official XML Schema and DTD specs further divide this into + * two modules: + * - Block Presentation (hr) + * - Inline Presentation (b, big, i, small, sub, sup, tt) + * We have chosen not to heed this distinction, as content_sets + * provides satisfactory disambiguation. + */ +class HTMLPurifier_HTMLModule_Presentation extends HTMLPurifier_HTMLModule +{ + + public $name = 'Presentation'; + + public function setup($config) { + $this->addElement('hr', 'Block', 'Empty', 'Common'); + $this->addElement('sub', 'Inline', 'Inline', 'Common'); + $this->addElement('sup', 'Inline', 'Inline', 'Common'); + $b = $this->addElement('b', 'Inline', 'Inline', 'Common'); + $b->formatting = true; + $big = $this->addElement('big', 'Inline', 'Inline', 'Common'); + $big->formatting = true; + $i = $this->addElement('i', 'Inline', 'Inline', 'Common'); + $i->formatting = true; + $small = $this->addElement('small', 'Inline', 'Inline', 'Common'); + $small->formatting = true; + $tt = $this->addElement('tt', 'Inline', 'Inline', 'Common'); + $tt->formatting = true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Proprietary.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Proprietary.php new file mode 100644 index 000000000..dd36a3de0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Proprietary.php @@ -0,0 +1,33 @@ +<?php + +/** + * Module defines proprietary tags and attributes in HTML. + * @warning If this module is enabled, standards-compliance is off! + */ +class HTMLPurifier_HTMLModule_Proprietary extends HTMLPurifier_HTMLModule +{ + + public $name = 'Proprietary'; + + public function setup($config) { + + $this->addElement('marquee', 'Inline', 'Flow', 'Common', + array( + 'direction' => 'Enum#left,right,up,down', + 'behavior' => 'Enum#alternate', + 'width' => 'Length', + 'height' => 'Length', + 'scrolldelay' => 'Number', + 'scrollamount' => 'Number', + 'loop' => 'Number', + 'bgcolor' => 'Color', + 'hspace' => 'Pixels', + 'vspace' => 'Pixels', + ) + ); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Ruby.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Ruby.php new file mode 100644 index 000000000..b26a0a30a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Ruby.php @@ -0,0 +1,27 @@ +<?php + +/** + * XHTML 1.1 Ruby Annotation Module, defines elements that indicate + * short runs of text alongside base text for annotation or pronounciation. + */ +class HTMLPurifier_HTMLModule_Ruby extends HTMLPurifier_HTMLModule +{ + + public $name = 'Ruby'; + + public function setup($config) { + $this->addElement('ruby', 'Inline', + 'Custom: ((rb, (rt | (rp, rt, rp))) | (rbc, rtc, rtc?))', + 'Common'); + $this->addElement('rbc', false, 'Required: rb', 'Common'); + $this->addElement('rtc', false, 'Required: rt', 'Common'); + $rb = $this->addElement('rb', false, 'Inline', 'Common'); + $rb->excludes = array('ruby' => true); + $rt = $this->addElement('rt', false, 'Inline', 'Common', array('rbspan' => 'Number')); + $rt->excludes = array('ruby' => true); + $this->addElement('rp', false, 'Optional: #PCDATA', 'Common'); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/SafeEmbed.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/SafeEmbed.php new file mode 100644 index 000000000..9f3758a32 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/SafeEmbed.php @@ -0,0 +1,34 @@ +<?php + +/** + * A "safe" embed module. See SafeObject. This is a proprietary element. + */ +class HTMLPurifier_HTMLModule_SafeEmbed extends HTMLPurifier_HTMLModule +{ + + public $name = 'SafeEmbed'; + + public function setup($config) { + + $max = $config->get('HTML.MaxImgLength'); + $embed = $this->addElement( + 'embed', 'Inline', 'Empty', 'Common', + array( + 'src*' => 'URI#embedded', + 'type' => 'Enum#application/x-shockwave-flash', + 'width' => 'Pixels#' . $max, + 'height' => 'Pixels#' . $max, + 'allowscriptaccess' => 'Enum#never', + 'allownetworking' => 'Enum#internal', + 'flashvars' => 'Text', + 'wmode' => 'Enum#window,transparent,opaque', + 'name' => 'ID', + ) + ); + $embed->attr_transform_post[] = new HTMLPurifier_AttrTransform_SafeEmbed(); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/SafeObject.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/SafeObject.php new file mode 100644 index 000000000..00da342ef --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/SafeObject.php @@ -0,0 +1,52 @@ +<?php + +/** + * A "safe" object module. In theory, objects permitted by this module will + * be safe, and untrusted users can be allowed to embed arbitrary flash objects + * (maybe other types too, but only Flash is supported as of right now). + * Highly experimental. + */ +class HTMLPurifier_HTMLModule_SafeObject extends HTMLPurifier_HTMLModule +{ + + public $name = 'SafeObject'; + + public function setup($config) { + + // These definitions are not intrinsically safe: the attribute transforms + // are a vital part of ensuring safety. + + $max = $config->get('HTML.MaxImgLength'); + $object = $this->addElement( + 'object', + 'Inline', + 'Optional: param | Flow | #PCDATA', + 'Common', + array( + // While technically not required by the spec, we're forcing + // it to this value. + 'type' => 'Enum#application/x-shockwave-flash', + 'width' => 'Pixels#' . $max, + 'height' => 'Pixels#' . $max, + 'data' => 'URI#embedded', + 'codebase' => new HTMLPurifier_AttrDef_Enum(array( + 'http://download.macromedia.com/pub/shockwave/cabs/flash/swflash.cab#version=6,0,40,0')), + ) + ); + $object->attr_transform_post[] = new HTMLPurifier_AttrTransform_SafeObject(); + + $param = $this->addElement('param', false, 'Empty', false, + array( + 'id' => 'ID', + 'name*' => 'Text', + 'value' => 'Text' + ) + ); + $param->attr_transform_post[] = new HTMLPurifier_AttrTransform_SafeParam(); + $this->info_injector[] = 'SafeObject'; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Scripting.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Scripting.php new file mode 100644 index 000000000..cecdea6c3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Scripting.php @@ -0,0 +1,54 @@ +<?php + +/* + +WARNING: THIS MODULE IS EXTREMELY DANGEROUS AS IT ENABLES INLINE SCRIPTING +INSIDE HTML PURIFIER DOCUMENTS. USE ONLY WITH TRUSTED USER INPUT!!! + +*/ + +/** + * XHTML 1.1 Scripting module, defines elements that are used to contain + * information pertaining to executable scripts or the lack of support + * for executable scripts. + * @note This module does not contain inline scripting elements + */ +class HTMLPurifier_HTMLModule_Scripting extends HTMLPurifier_HTMLModule +{ + public $name = 'Scripting'; + public $elements = array('script', 'noscript'); + public $content_sets = array('Block' => 'script | noscript', 'Inline' => 'script | noscript'); + public $safe = false; + + public function setup($config) { + // TODO: create custom child-definition for noscript that + // auto-wraps stray #PCDATA in a similar manner to + // blockquote's custom definition (we would use it but + // blockquote's contents are optional while noscript's contents + // are required) + + // TODO: convert this to new syntax, main problem is getting + // both content sets working + + // In theory, this could be safe, but I don't see any reason to + // allow it. + $this->info['noscript'] = new HTMLPurifier_ElementDef(); + $this->info['noscript']->attr = array( 0 => array('Common') ); + $this->info['noscript']->content_model = 'Heading | List | Block'; + $this->info['noscript']->content_model_type = 'required'; + + $this->info['script'] = new HTMLPurifier_ElementDef(); + $this->info['script']->attr = array( + 'defer' => new HTMLPurifier_AttrDef_Enum(array('defer')), + 'src' => new HTMLPurifier_AttrDef_URI(true), + 'type' => new HTMLPurifier_AttrDef_Enum(array('text/javascript')) + ); + $this->info['script']->content_model = '#PCDATA'; + $this->info['script']->content_model_type = 'optional'; + $this->info['script']->attr_transform_pre['type'] = + $this->info['script']->attr_transform_post['type'] = + new HTMLPurifier_AttrTransform_ScriptRequired(); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/StyleAttribute.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/StyleAttribute.php new file mode 100644 index 000000000..eb78464cc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/StyleAttribute.php @@ -0,0 +1,24 @@ +<?php + +/** + * XHTML 1.1 Edit Module, defines editing-related elements. Text Extension + * Module. + */ +class HTMLPurifier_HTMLModule_StyleAttribute extends HTMLPurifier_HTMLModule +{ + + public $name = 'StyleAttribute'; + public $attr_collections = array( + // The inclusion routine differs from the Abstract Modules but + // is in line with the DTD and XML Schemas. + 'Style' => array('style' => false), // see constructor + 'Core' => array(0 => array('Style')) + ); + + public function setup($config) { + $this->attr_collections['Style']['style'] = new HTMLPurifier_AttrDef_CSS(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tables.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tables.php new file mode 100644 index 000000000..45c42bb3e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tables.php @@ -0,0 +1,69 @@ +<?php + +/** + * XHTML 1.1 Tables Module, fully defines accessible table elements. + */ +class HTMLPurifier_HTMLModule_Tables extends HTMLPurifier_HTMLModule +{ + + public $name = 'Tables'; + + public function setup($config) { + + $this->addElement('caption', false, 'Inline', 'Common'); + + $this->addElement('table', 'Block', + new HTMLPurifier_ChildDef_Table(), 'Common', + array( + 'border' => 'Pixels', + 'cellpadding' => 'Length', + 'cellspacing' => 'Length', + 'frame' => 'Enum#void,above,below,hsides,lhs,rhs,vsides,box,border', + 'rules' => 'Enum#none,groups,rows,cols,all', + 'summary' => 'Text', + 'width' => 'Length' + ) + ); + + // common attributes + $cell_align = array( + 'align' => 'Enum#left,center,right,justify,char', + 'charoff' => 'Length', + 'valign' => 'Enum#top,middle,bottom,baseline', + ); + + $cell_t = array_merge( + array( + 'abbr' => 'Text', + 'colspan' => 'Number', + 'rowspan' => 'Number', + // Apparently, as of HTML5 this attribute only applies + // to 'th' elements. + 'scope' => 'Enum#row,col,rowgroup,colgroup', + ), + $cell_align + ); + $this->addElement('td', false, 'Flow', 'Common', $cell_t); + $this->addElement('th', false, 'Flow', 'Common', $cell_t); + + $this->addElement('tr', false, 'Required: td | th', 'Common', $cell_align); + + $cell_col = array_merge( + array( + 'span' => 'Number', + 'width' => 'MultiLength', + ), + $cell_align + ); + $this->addElement('col', false, 'Empty', 'Common', $cell_col); + $this->addElement('colgroup', false, 'Optional: col', 'Common', $cell_col); + + $this->addElement('tbody', false, 'Required: tr', 'Common', $cell_align); + $this->addElement('thead', false, 'Required: tr', 'Common', $cell_align); + $this->addElement('tfoot', false, 'Required: tr', 'Common', $cell_align); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Target.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Target.php new file mode 100644 index 000000000..2b844ecc4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Target.php @@ -0,0 +1,23 @@ +<?php + +/** + * XHTML 1.1 Target Module, defines target attribute in link elements. + */ +class HTMLPurifier_HTMLModule_Target extends HTMLPurifier_HTMLModule +{ + + public $name = 'Target'; + + public function setup($config) { + $elements = array('a'); + foreach ($elements as $name) { + $e = $this->addBlankElement($name); + $e->attr = array( + 'target' => new HTMLPurifier_AttrDef_HTML_FrameTarget() + ); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/TargetBlank.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/TargetBlank.php new file mode 100644 index 000000000..e1305ec5d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/TargetBlank.php @@ -0,0 +1,19 @@ +<?php + +/** + * Module adds the target=blank attribute transformation to a tags. It + * is enabled by HTML.TargetBlank + */ +class HTMLPurifier_HTMLModule_TargetBlank extends HTMLPurifier_HTMLModule +{ + + public $name = 'TargetBlank'; + + public function setup($config) { + $a = $this->addBlankElement('a'); + $a->attr_transform_post[] = new HTMLPurifier_AttrTransform_TargetBlank(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Text.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Text.php new file mode 100644 index 000000000..ae77c7188 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Text.php @@ -0,0 +1,71 @@ +<?php + +/** + * XHTML 1.1 Text Module, defines basic text containers. Core Module. + * @note In the normative XML Schema specification, this module + * is further abstracted into the following modules: + * - Block Phrasal (address, blockquote, pre, h1, h2, h3, h4, h5, h6) + * - Block Structural (div, p) + * - Inline Phrasal (abbr, acronym, cite, code, dfn, em, kbd, q, samp, strong, var) + * - Inline Structural (br, span) + * This module, functionally, does not distinguish between these + * sub-modules, but the code is internally structured to reflect + * these distinctions. + */ +class HTMLPurifier_HTMLModule_Text extends HTMLPurifier_HTMLModule +{ + + public $name = 'Text'; + public $content_sets = array( + 'Flow' => 'Heading | Block | Inline' + ); + + public function setup($config) { + + // Inline Phrasal ------------------------------------------------- + $this->addElement('abbr', 'Inline', 'Inline', 'Common'); + $this->addElement('acronym', 'Inline', 'Inline', 'Common'); + $this->addElement('cite', 'Inline', 'Inline', 'Common'); + $this->addElement('dfn', 'Inline', 'Inline', 'Common'); + $this->addElement('kbd', 'Inline', 'Inline', 'Common'); + $this->addElement('q', 'Inline', 'Inline', 'Common', array('cite' => 'URI')); + $this->addElement('samp', 'Inline', 'Inline', 'Common'); + $this->addElement('var', 'Inline', 'Inline', 'Common'); + + $em = $this->addElement('em', 'Inline', 'Inline', 'Common'); + $em->formatting = true; + + $strong = $this->addElement('strong', 'Inline', 'Inline', 'Common'); + $strong->formatting = true; + + $code = $this->addElement('code', 'Inline', 'Inline', 'Common'); + $code->formatting = true; + + // Inline Structural ---------------------------------------------- + $this->addElement('span', 'Inline', 'Inline', 'Common'); + $this->addElement('br', 'Inline', 'Empty', 'Core'); + + // Block Phrasal -------------------------------------------------- + $this->addElement('address', 'Block', 'Inline', 'Common'); + $this->addElement('blockquote', 'Block', 'Optional: Heading | Block | List', 'Common', array('cite' => 'URI') ); + $pre = $this->addElement('pre', 'Block', 'Inline', 'Common'); + $pre->excludes = $this->makeLookup( + 'img', 'big', 'small', 'object', 'applet', 'font', 'basefont' ); + $this->addElement('h1', 'Heading', 'Inline', 'Common'); + $this->addElement('h2', 'Heading', 'Inline', 'Common'); + $this->addElement('h3', 'Heading', 'Inline', 'Common'); + $this->addElement('h4', 'Heading', 'Inline', 'Common'); + $this->addElement('h5', 'Heading', 'Inline', 'Common'); + $this->addElement('h6', 'Heading', 'Inline', 'Common'); + + // Block Structural ----------------------------------------------- + $p = $this->addElement('p', 'Block', 'Inline', 'Common'); + $p->autoclose = array_flip(array("address", "blockquote", "center", "dir", "div", "dl", "fieldset", "ol", "p", "ul")); + + $this->addElement('div', 'Block', 'Flow', 'Common'); + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy.php new file mode 100644 index 000000000..21783f18e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy.php @@ -0,0 +1,207 @@ +<?php + +/** + * Abstract class for a set of proprietary modules that clean up (tidy) + * poorly written HTML. + * @todo Figure out how to protect some of these methods/properties + */ +class HTMLPurifier_HTMLModule_Tidy extends HTMLPurifier_HTMLModule +{ + + /** + * List of supported levels. Index zero is a special case "no fixes" + * level. + */ + public $levels = array(0 => 'none', 'light', 'medium', 'heavy'); + + /** + * Default level to place all fixes in. Disabled by default + */ + public $defaultLevel = null; + + /** + * Lists of fixes used by getFixesForLevel(). Format is: + * HTMLModule_Tidy->fixesForLevel[$level] = array('fix-1', 'fix-2'); + */ + public $fixesForLevel = array( + 'light' => array(), + 'medium' => array(), + 'heavy' => array() + ); + + /** + * Lazy load constructs the module by determining the necessary + * fixes to create and then delegating to the populate() function. + * @todo Wildcard matching and error reporting when an added or + * subtracted fix has no effect. + */ + public function setup($config) { + + // create fixes, initialize fixesForLevel + $fixes = $this->makeFixes(); + $this->makeFixesForLevel($fixes); + + // figure out which fixes to use + $level = $config->get('HTML.TidyLevel'); + $fixes_lookup = $this->getFixesForLevel($level); + + // get custom fix declarations: these need namespace processing + $add_fixes = $config->get('HTML.TidyAdd'); + $remove_fixes = $config->get('HTML.TidyRemove'); + + foreach ($fixes as $name => $fix) { + // needs to be refactored a little to implement globbing + if ( + isset($remove_fixes[$name]) || + (!isset($add_fixes[$name]) && !isset($fixes_lookup[$name])) + ) { + unset($fixes[$name]); + } + } + + // populate this module with necessary fixes + $this->populate($fixes); + + } + + /** + * Retrieves all fixes per a level, returning fixes for that specific + * level as well as all levels below it. + * @param $level String level identifier, see $levels for valid values + * @return Lookup up table of fixes + */ + public function getFixesForLevel($level) { + if ($level == $this->levels[0]) { + return array(); + } + $activated_levels = array(); + for ($i = 1, $c = count($this->levels); $i < $c; $i++) { + $activated_levels[] = $this->levels[$i]; + if ($this->levels[$i] == $level) break; + } + if ($i == $c) { + trigger_error( + 'Tidy level ' . htmlspecialchars($level) . ' not recognized', + E_USER_WARNING + ); + return array(); + } + $ret = array(); + foreach ($activated_levels as $level) { + foreach ($this->fixesForLevel[$level] as $fix) { + $ret[$fix] = true; + } + } + return $ret; + } + + /** + * Dynamically populates the $fixesForLevel member variable using + * the fixes array. It may be custom overloaded, used in conjunction + * with $defaultLevel, or not used at all. + */ + public function makeFixesForLevel($fixes) { + if (!isset($this->defaultLevel)) return; + if (!isset($this->fixesForLevel[$this->defaultLevel])) { + trigger_error( + 'Default level ' . $this->defaultLevel . ' does not exist', + E_USER_ERROR + ); + return; + } + $this->fixesForLevel[$this->defaultLevel] = array_keys($fixes); + } + + /** + * Populates the module with transforms and other special-case code + * based on a list of fixes passed to it + * @param $lookup Lookup table of fixes to activate + */ + public function populate($fixes) { + foreach ($fixes as $name => $fix) { + // determine what the fix is for + list($type, $params) = $this->getFixType($name); + switch ($type) { + case 'attr_transform_pre': + case 'attr_transform_post': + $attr = $params['attr']; + if (isset($params['element'])) { + $element = $params['element']; + if (empty($this->info[$element])) { + $e = $this->addBlankElement($element); + } else { + $e = $this->info[$element]; + } + } else { + $type = "info_$type"; + $e = $this; + } + // PHP does some weird parsing when I do + // $e->$type[$attr], so I have to assign a ref. + $f =& $e->$type; + $f[$attr] = $fix; + break; + case 'tag_transform': + $this->info_tag_transform[$params['element']] = $fix; + break; + case 'child': + case 'content_model_type': + $element = $params['element']; + if (empty($this->info[$element])) { + $e = $this->addBlankElement($element); + } else { + $e = $this->info[$element]; + } + $e->$type = $fix; + break; + default: + trigger_error("Fix type $type not supported", E_USER_ERROR); + break; + } + } + } + + /** + * Parses a fix name and determines what kind of fix it is, as well + * as other information defined by the fix + * @param $name String name of fix + * @return array(string $fix_type, array $fix_parameters) + * @note $fix_parameters is type dependant, see populate() for usage + * of these parameters + */ + public function getFixType($name) { + // parse it + $property = $attr = null; + if (strpos($name, '#') !== false) list($name, $property) = explode('#', $name); + if (strpos($name, '@') !== false) list($name, $attr) = explode('@', $name); + + // figure out the parameters + $params = array(); + if ($name !== '') $params['element'] = $name; + if (!is_null($attr)) $params['attr'] = $attr; + + // special case: attribute transform + if (!is_null($attr)) { + if (is_null($property)) $property = 'pre'; + $type = 'attr_transform_' . $property; + return array($type, $params); + } + + // special case: tag transform + if (is_null($property)) { + return array('tag_transform', $params); + } + + return array($property, $params); + + } + + /** + * Defines all fixes the module will perform in a compact + * associative array of fix name to fix implementation. + */ + public function makeFixes() {} + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Name.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Name.php new file mode 100644 index 000000000..61ff85ce2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Name.php @@ -0,0 +1,24 @@ +<?php + +/** + * Name is deprecated, but allowed in strict doctypes, so onl + */ +class HTMLPurifier_HTMLModule_Tidy_Name extends HTMLPurifier_HTMLModule_Tidy +{ + public $name = 'Tidy_Name'; + public $defaultLevel = 'heavy'; + public function makeFixes() { + + $r = array(); + + // @name for img, a ----------------------------------------------- + // Technically, it's allowed even on strict, so we allow authors to use + // it. However, it's deprecated in future versions of XHTML. + $r['img@name'] = + $r['a@name'] = new HTMLPurifier_AttrTransform_Name(); + + return $r; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Proprietary.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Proprietary.php new file mode 100644 index 000000000..14c15c4a0 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Proprietary.php @@ -0,0 +1,24 @@ +<?php + +class HTMLPurifier_HTMLModule_Tidy_Proprietary extends HTMLPurifier_HTMLModule_Tidy +{ + + public $name = 'Tidy_Proprietary'; + public $defaultLevel = 'light'; + + public function makeFixes() { + $r = array(); + $r['table@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['td@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['th@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['tr@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['thead@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['tfoot@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['tbody@background'] = new HTMLPurifier_AttrTransform_Background(); + $r['table@height'] = new HTMLPurifier_AttrTransform_Length('height'); + return $r; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Strict.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Strict.php new file mode 100644 index 000000000..c73dc3c4d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Strict.php @@ -0,0 +1,21 @@ +<?php + +class HTMLPurifier_HTMLModule_Tidy_Strict extends HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4 +{ + public $name = 'Tidy_Strict'; + public $defaultLevel = 'light'; + + public function makeFixes() { + $r = parent::makeFixes(); + $r['blockquote#content_model_type'] = 'strictblockquote'; + return $r; + } + + public $defines_child_def = true; + public function getChildDef($def) { + if ($def->content_model_type != 'strictblockquote') return parent::getChildDef($def); + return new HTMLPurifier_ChildDef_StrictBlockquote($def->content_model); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Transitional.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Transitional.php new file mode 100644 index 000000000..9960b1dd1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/Transitional.php @@ -0,0 +1,9 @@ +<?php + +class HTMLPurifier_HTMLModule_Tidy_Transitional extends HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4 +{ + public $name = 'Tidy_Transitional'; + public $defaultLevel = 'heavy'; +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/XHTML.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/XHTML.php new file mode 100644 index 000000000..db5a378e5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/XHTML.php @@ -0,0 +1,17 @@ +<?php + +class HTMLPurifier_HTMLModule_Tidy_XHTML extends HTMLPurifier_HTMLModule_Tidy +{ + + public $name = 'Tidy_XHTML'; + public $defaultLevel = 'medium'; + + public function makeFixes() { + $r = array(); + $r['@lang'] = new HTMLPurifier_AttrTransform_Lang(); + return $r; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php new file mode 100644 index 000000000..02e943813 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/Tidy/XHTMLAndHTML4.php @@ -0,0 +1,161 @@ +<?php + +class HTMLPurifier_HTMLModule_Tidy_XHTMLAndHTML4 extends HTMLPurifier_HTMLModule_Tidy +{ + + public function makeFixes() { + + $r = array(); + + // == deprecated tag transforms =================================== + + $r['font'] = new HTMLPurifier_TagTransform_Font(); + $r['menu'] = new HTMLPurifier_TagTransform_Simple('ul'); + $r['dir'] = new HTMLPurifier_TagTransform_Simple('ul'); + $r['center'] = new HTMLPurifier_TagTransform_Simple('div', 'text-align:center;'); + $r['u'] = new HTMLPurifier_TagTransform_Simple('span', 'text-decoration:underline;'); + $r['s'] = new HTMLPurifier_TagTransform_Simple('span', 'text-decoration:line-through;'); + $r['strike'] = new HTMLPurifier_TagTransform_Simple('span', 'text-decoration:line-through;'); + + // == deprecated attribute transforms ============================= + + $r['caption@align'] = + new HTMLPurifier_AttrTransform_EnumToCSS('align', array( + // we're following IE's behavior, not Firefox's, due + // to the fact that no one supports caption-side:right, + // W3C included (with CSS 2.1). This is a slightly + // unreasonable attribute! + 'left' => 'text-align:left;', + 'right' => 'text-align:right;', + 'top' => 'caption-side:top;', + 'bottom' => 'caption-side:bottom;' // not supported by IE + )); + + // @align for img ------------------------------------------------- + $r['img@align'] = + new HTMLPurifier_AttrTransform_EnumToCSS('align', array( + 'left' => 'float:left;', + 'right' => 'float:right;', + 'top' => 'vertical-align:top;', + 'middle' => 'vertical-align:middle;', + 'bottom' => 'vertical-align:baseline;', + )); + + // @align for table ----------------------------------------------- + $r['table@align'] = + new HTMLPurifier_AttrTransform_EnumToCSS('align', array( + 'left' => 'float:left;', + 'center' => 'margin-left:auto;margin-right:auto;', + 'right' => 'float:right;' + )); + + // @align for hr ----------------------------------------------- + $r['hr@align'] = + new HTMLPurifier_AttrTransform_EnumToCSS('align', array( + // we use both text-align and margin because these work + // for different browsers (IE and Firefox, respectively) + // and the melange makes for a pretty cross-compatible + // solution + 'left' => 'margin-left:0;margin-right:auto;text-align:left;', + 'center' => 'margin-left:auto;margin-right:auto;text-align:center;', + 'right' => 'margin-left:auto;margin-right:0;text-align:right;' + )); + + // @align for h1, h2, h3, h4, h5, h6, p, div ---------------------- + // {{{ + $align_lookup = array(); + $align_values = array('left', 'right', 'center', 'justify'); + foreach ($align_values as $v) $align_lookup[$v] = "text-align:$v;"; + // }}} + $r['h1@align'] = + $r['h2@align'] = + $r['h3@align'] = + $r['h4@align'] = + $r['h5@align'] = + $r['h6@align'] = + $r['p@align'] = + $r['div@align'] = + new HTMLPurifier_AttrTransform_EnumToCSS('align', $align_lookup); + + // @bgcolor for table, tr, td, th --------------------------------- + $r['table@bgcolor'] = + $r['td@bgcolor'] = + $r['th@bgcolor'] = + new HTMLPurifier_AttrTransform_BgColor(); + + // @border for img ------------------------------------------------ + $r['img@border'] = new HTMLPurifier_AttrTransform_Border(); + + // @clear for br -------------------------------------------------- + $r['br@clear'] = + new HTMLPurifier_AttrTransform_EnumToCSS('clear', array( + 'left' => 'clear:left;', + 'right' => 'clear:right;', + 'all' => 'clear:both;', + 'none' => 'clear:none;', + )); + + // @height for td, th --------------------------------------------- + $r['td@height'] = + $r['th@height'] = + new HTMLPurifier_AttrTransform_Length('height'); + + // @hspace for img ------------------------------------------------ + $r['img@hspace'] = new HTMLPurifier_AttrTransform_ImgSpace('hspace'); + + // @noshade for hr ------------------------------------------------ + // this transformation is not precise but often good enough. + // different browsers use different styles to designate noshade + $r['hr@noshade'] = + new HTMLPurifier_AttrTransform_BoolToCSS( + 'noshade', + 'color:#808080;background-color:#808080;border:0;' + ); + + // @nowrap for td, th --------------------------------------------- + $r['td@nowrap'] = + $r['th@nowrap'] = + new HTMLPurifier_AttrTransform_BoolToCSS( + 'nowrap', + 'white-space:nowrap;' + ); + + // @size for hr -------------------------------------------------- + $r['hr@size'] = new HTMLPurifier_AttrTransform_Length('size', 'height'); + + // @type for li, ol, ul ------------------------------------------- + // {{{ + $ul_types = array( + 'disc' => 'list-style-type:disc;', + 'square' => 'list-style-type:square;', + 'circle' => 'list-style-type:circle;' + ); + $ol_types = array( + '1' => 'list-style-type:decimal;', + 'i' => 'list-style-type:lower-roman;', + 'I' => 'list-style-type:upper-roman;', + 'a' => 'list-style-type:lower-alpha;', + 'A' => 'list-style-type:upper-alpha;' + ); + $li_types = $ul_types + $ol_types; + // }}} + + $r['ul@type'] = new HTMLPurifier_AttrTransform_EnumToCSS('type', $ul_types); + $r['ol@type'] = new HTMLPurifier_AttrTransform_EnumToCSS('type', $ol_types, true); + $r['li@type'] = new HTMLPurifier_AttrTransform_EnumToCSS('type', $li_types, true); + + // @vspace for img ------------------------------------------------ + $r['img@vspace'] = new HTMLPurifier_AttrTransform_ImgSpace('vspace'); + + // @width for hr, td, th ------------------------------------------ + $r['td@width'] = + $r['th@width'] = + $r['hr@width'] = new HTMLPurifier_AttrTransform_Length('width'); + + return $r; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/XMLCommonAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/XMLCommonAttributes.php new file mode 100644 index 000000000..9c0e03198 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModule/XMLCommonAttributes.php @@ -0,0 +1,14 @@ +<?php + +class HTMLPurifier_HTMLModule_XMLCommonAttributes extends HTMLPurifier_HTMLModule +{ + public $name = 'XMLCommonAttributes'; + + public $attr_collections = array( + 'Lang' => array( + 'xml:lang' => 'LanguageCode', + ) + ); +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/HTMLModuleManager.php b/lib/htmlpurifier/library/HTMLPurifier/HTMLModuleManager.php new file mode 100644 index 000000000..7a06fc022 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/HTMLModuleManager.php @@ -0,0 +1,415 @@ +<?php + +class HTMLPurifier_HTMLModuleManager +{ + + /** + * Instance of HTMLPurifier_DoctypeRegistry + */ + public $doctypes; + + /** + * Instance of current doctype + */ + public $doctype; + + /** + * Instance of HTMLPurifier_AttrTypes + */ + public $attrTypes; + + /** + * Active instances of modules for the specified doctype are + * indexed, by name, in this array. + */ + public $modules = array(); + + /** + * Array of recognized HTMLPurifier_Module instances, indexed by + * module's class name. This array is usually lazy loaded, but a + * user can overload a module by pre-emptively registering it. + */ + public $registeredModules = array(); + + /** + * List of extra modules that were added by the user using addModule(). + * These get unconditionally merged into the current doctype, whatever + * it may be. + */ + public $userModules = array(); + + /** + * Associative array of element name to list of modules that have + * definitions for the element; this array is dynamically filled. + */ + public $elementLookup = array(); + + /** List of prefixes we should use for registering small names */ + public $prefixes = array('HTMLPurifier_HTMLModule_'); + + public $contentSets; /**< Instance of HTMLPurifier_ContentSets */ + public $attrCollections; /**< Instance of HTMLPurifier_AttrCollections */ + + /** If set to true, unsafe elements and attributes will be allowed */ + public $trusted = false; + + public function __construct() { + + // editable internal objects + $this->attrTypes = new HTMLPurifier_AttrTypes(); + $this->doctypes = new HTMLPurifier_DoctypeRegistry(); + + // setup basic modules + $common = array( + 'CommonAttributes', 'Text', 'Hypertext', 'List', + 'Presentation', 'Edit', 'Bdo', 'Tables', 'Image', + 'StyleAttribute', + // Unsafe: + 'Scripting', 'Object', 'Forms', + // Sorta legacy, but present in strict: + 'Name', + ); + $transitional = array('Legacy', 'Target', 'Iframe'); + $xml = array('XMLCommonAttributes'); + $non_xml = array('NonXMLCommonAttributes'); + + // setup basic doctypes + $this->doctypes->register( + 'HTML 4.01 Transitional', false, + array_merge($common, $transitional, $non_xml), + array('Tidy_Transitional', 'Tidy_Proprietary'), + array(), + '-//W3C//DTD HTML 4.01 Transitional//EN', + 'http://www.w3.org/TR/html4/loose.dtd' + ); + + $this->doctypes->register( + 'HTML 4.01 Strict', false, + array_merge($common, $non_xml), + array('Tidy_Strict', 'Tidy_Proprietary', 'Tidy_Name'), + array(), + '-//W3C//DTD HTML 4.01//EN', + 'http://www.w3.org/TR/html4/strict.dtd' + ); + + $this->doctypes->register( + 'XHTML 1.0 Transitional', true, + array_merge($common, $transitional, $xml, $non_xml), + array('Tidy_Transitional', 'Tidy_XHTML', 'Tidy_Proprietary', 'Tidy_Name'), + array(), + '-//W3C//DTD XHTML 1.0 Transitional//EN', + 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd' + ); + + $this->doctypes->register( + 'XHTML 1.0 Strict', true, + array_merge($common, $xml, $non_xml), + array('Tidy_Strict', 'Tidy_XHTML', 'Tidy_Strict', 'Tidy_Proprietary', 'Tidy_Name'), + array(), + '-//W3C//DTD XHTML 1.0 Strict//EN', + 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd' + ); + + $this->doctypes->register( + 'XHTML 1.1', true, + // Iframe is a real XHTML 1.1 module, despite being + // "transitional"! + array_merge($common, $xml, array('Ruby', 'Iframe')), + array('Tidy_Strict', 'Tidy_XHTML', 'Tidy_Proprietary', 'Tidy_Strict', 'Tidy_Name'), // Tidy_XHTML1_1 + array(), + '-//W3C//DTD XHTML 1.1//EN', + 'http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd' + ); + + } + + /** + * Registers a module to the recognized module list, useful for + * overloading pre-existing modules. + * @param $module Mixed: string module name, with or without + * HTMLPurifier_HTMLModule prefix, or instance of + * subclass of HTMLPurifier_HTMLModule. + * @param $overload Boolean whether or not to overload previous modules. + * If this is not set, and you do overload a module, + * HTML Purifier will complain with a warning. + * @note This function will not call autoload, you must instantiate + * (and thus invoke) autoload outside the method. + * @note If a string is passed as a module name, different variants + * will be tested in this order: + * - Check for HTMLPurifier_HTMLModule_$name + * - Check all prefixes with $name in order they were added + * - Check for literal object name + * - Throw fatal error + * If your object name collides with an internal class, specify + * your module manually. All modules must have been included + * externally: registerModule will not perform inclusions for you! + */ + public function registerModule($module, $overload = false) { + if (is_string($module)) { + // attempt to load the module + $original_module = $module; + $ok = false; + foreach ($this->prefixes as $prefix) { + $module = $prefix . $original_module; + if (class_exists($module)) { + $ok = true; + break; + } + } + if (!$ok) { + $module = $original_module; + if (!class_exists($module)) { + trigger_error($original_module . ' module does not exist', + E_USER_ERROR); + return; + } + } + $module = new $module(); + } + if (empty($module->name)) { + trigger_error('Module instance of ' . get_class($module) . ' must have name'); + return; + } + if (!$overload && isset($this->registeredModules[$module->name])) { + trigger_error('Overloading ' . $module->name . ' without explicit overload parameter', E_USER_WARNING); + } + $this->registeredModules[$module->name] = $module; + } + + /** + * Adds a module to the current doctype by first registering it, + * and then tacking it on to the active doctype + */ + public function addModule($module) { + $this->registerModule($module); + if (is_object($module)) $module = $module->name; + $this->userModules[] = $module; + } + + /** + * Adds a class prefix that registerModule() will use to resolve a + * string name to a concrete class + */ + public function addPrefix($prefix) { + $this->prefixes[] = $prefix; + } + + /** + * Performs processing on modules, after being called you may + * use getElement() and getElements() + * @param $config Instance of HTMLPurifier_Config + */ + public function setup($config) { + + $this->trusted = $config->get('HTML.Trusted'); + + // generate + $this->doctype = $this->doctypes->make($config); + $modules = $this->doctype->modules; + + // take out the default modules that aren't allowed + $lookup = $config->get('HTML.AllowedModules'); + $special_cases = $config->get('HTML.CoreModules'); + + if (is_array($lookup)) { + foreach ($modules as $k => $m) { + if (isset($special_cases[$m])) continue; + if (!isset($lookup[$m])) unset($modules[$k]); + } + } + + // custom modules + if ($config->get('HTML.Proprietary')) { + $modules[] = 'Proprietary'; + } + if ($config->get('HTML.SafeObject')) { + $modules[] = 'SafeObject'; + } + if ($config->get('HTML.SafeEmbed')) { + $modules[] = 'SafeEmbed'; + } + if ($config->get('HTML.Nofollow')) { + $modules[] = 'Nofollow'; + } + if ($config->get('HTML.TargetBlank')) { + $modules[] = 'TargetBlank'; + } + + // merge in custom modules + $modules = array_merge($modules, $this->userModules); + + foreach ($modules as $module) { + $this->processModule($module); + $this->modules[$module]->setup($config); + } + + foreach ($this->doctype->tidyModules as $module) { + $this->processModule($module); + $this->modules[$module]->setup($config); + } + + // prepare any injectors + foreach ($this->modules as $module) { + $n = array(); + foreach ($module->info_injector as $i => $injector) { + if (!is_object($injector)) { + $class = "HTMLPurifier_Injector_$injector"; + $injector = new $class; + } + $n[$injector->name] = $injector; + } + $module->info_injector = $n; + } + + // setup lookup table based on all valid modules + foreach ($this->modules as $module) { + foreach ($module->info as $name => $def) { + if (!isset($this->elementLookup[$name])) { + $this->elementLookup[$name] = array(); + } + $this->elementLookup[$name][] = $module->name; + } + } + + // note the different choice + $this->contentSets = new HTMLPurifier_ContentSets( + // content set assembly deals with all possible modules, + // not just ones deemed to be "safe" + $this->modules + ); + $this->attrCollections = new HTMLPurifier_AttrCollections( + $this->attrTypes, + // there is no way to directly disable a global attribute, + // but using AllowedAttributes or simply not including + // the module in your custom doctype should be sufficient + $this->modules + ); + } + + /** + * Takes a module and adds it to the active module collection, + * registering it if necessary. + */ + public function processModule($module) { + if (!isset($this->registeredModules[$module]) || is_object($module)) { + $this->registerModule($module); + } + $this->modules[$module] = $this->registeredModules[$module]; + } + + /** + * Retrieves merged element definitions. + * @return Array of HTMLPurifier_ElementDef + */ + public function getElements() { + + $elements = array(); + foreach ($this->modules as $module) { + if (!$this->trusted && !$module->safe) continue; + foreach ($module->info as $name => $v) { + if (isset($elements[$name])) continue; + $elements[$name] = $this->getElement($name); + } + } + + // remove dud elements, this happens when an element that + // appeared to be safe actually wasn't + foreach ($elements as $n => $v) { + if ($v === false) unset($elements[$n]); + } + + return $elements; + + } + + /** + * Retrieves a single merged element definition + * @param $name Name of element + * @param $trusted Boolean trusted overriding parameter: set to true + * if you want the full version of an element + * @return Merged HTMLPurifier_ElementDef + * @note You may notice that modules are getting iterated over twice (once + * in getElements() and once here). This + * is because + */ + public function getElement($name, $trusted = null) { + + if (!isset($this->elementLookup[$name])) { + return false; + } + + // setup global state variables + $def = false; + if ($trusted === null) $trusted = $this->trusted; + + // iterate through each module that has registered itself to this + // element + foreach($this->elementLookup[$name] as $module_name) { + + $module = $this->modules[$module_name]; + + // refuse to create/merge from a module that is deemed unsafe-- + // pretend the module doesn't exist--when trusted mode is not on. + if (!$trusted && !$module->safe) { + continue; + } + + // clone is used because, ideally speaking, the original + // definition should not be modified. Usually, this will + // make no difference, but for consistency's sake + $new_def = clone $module->info[$name]; + + if (!$def && $new_def->standalone) { + $def = $new_def; + } elseif ($def) { + // This will occur even if $new_def is standalone. In practice, + // this will usually result in a full replacement. + $def->mergeIn($new_def); + } else { + // :TODO: + // non-standalone definitions that don't have a standalone + // to merge into could be deferred to the end + // HOWEVER, it is perfectly valid for a non-standalone + // definition to lack a standalone definition, even + // after all processing: this allows us to safely + // specify extra attributes for elements that may not be + // enabled all in one place. In particular, this might + // be the case for trusted elements. WARNING: care must + // be taken that the /extra/ definitions are all safe. + continue; + } + + // attribute value expansions + $this->attrCollections->performInclusions($def->attr); + $this->attrCollections->expandIdentifiers($def->attr, $this->attrTypes); + + // descendants_are_inline, for ChildDef_Chameleon + if (is_string($def->content_model) && + strpos($def->content_model, 'Inline') !== false) { + if ($name != 'del' && $name != 'ins') { + // this is for you, ins/del + $def->descendants_are_inline = true; + } + } + + $this->contentSets->generateChildDef($def, $module); + } + + // This can occur if there is a blank definition, but no base to + // mix it in with + if (!$def) return false; + + // add information on required attributes + foreach ($def->attr as $attr_name => $attr_def) { + if ($attr_def->required) { + $def->required_attr[] = $attr_name; + } + } + + return $def; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/IDAccumulator.php b/lib/htmlpurifier/library/HTMLPurifier/IDAccumulator.php new file mode 100644 index 000000000..73215295a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/IDAccumulator.php @@ -0,0 +1,53 @@ +<?php + +/** + * Component of HTMLPurifier_AttrContext that accumulates IDs to prevent dupes + * @note In Slashdot-speak, dupe means duplicate. + * @note The default constructor does not accept $config or $context objects: + * use must use the static build() factory method to perform initialization. + */ +class HTMLPurifier_IDAccumulator +{ + + /** + * Lookup table of IDs we've accumulated. + * @public + */ + public $ids = array(); + + /** + * Builds an IDAccumulator, also initializing the default blacklist + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return Fully initialized HTMLPurifier_IDAccumulator + */ + public static function build($config, $context) { + $id_accumulator = new HTMLPurifier_IDAccumulator(); + $id_accumulator->load($config->get('Attr.IDBlacklist')); + return $id_accumulator; + } + + /** + * Add an ID to the lookup table. + * @param $id ID to be added. + * @return Bool status, true if success, false if there's a dupe + */ + public function add($id) { + if (isset($this->ids[$id])) return false; + return $this->ids[$id] = true; + } + + /** + * Load a list of IDs into the lookup table + * @param $array_of_ids Array of IDs to load + * @note This function doesn't care about duplicates + */ + public function load($array_of_ids) { + foreach ($array_of_ids as $id) { + $this->ids[$id] = true; + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector.php b/lib/htmlpurifier/library/HTMLPurifier/Injector.php new file mode 100644 index 000000000..5922f8130 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector.php @@ -0,0 +1,239 @@ +<?php + +/** + * Injects tokens into the document while parsing for well-formedness. + * This enables "formatter-like" functionality such as auto-paragraphing, + * smiley-ification and linkification to take place. + * + * A note on how handlers create changes; this is done by assigning a new + * value to the $token reference. These values can take a variety of forms and + * are best described HTMLPurifier_Strategy_MakeWellFormed->processToken() + * documentation. + * + * @todo Allow injectors to request a re-run on their output. This + * would help if an operation is recursive. + */ +abstract class HTMLPurifier_Injector +{ + + /** + * Advisory name of injector, this is for friendly error messages + */ + public $name; + + /** + * Instance of HTMLPurifier_HTMLDefinition + */ + protected $htmlDefinition; + + /** + * Reference to CurrentNesting variable in Context. This is an array + * list of tokens that we are currently "inside" + */ + protected $currentNesting; + + /** + * Reference to InputTokens variable in Context. This is an array + * list of the input tokens that are being processed. + */ + protected $inputTokens; + + /** + * Reference to InputIndex variable in Context. This is an integer + * array index for $this->inputTokens that indicates what token + * is currently being processed. + */ + protected $inputIndex; + + /** + * Array of elements and attributes this injector creates and therefore + * need to be allowed by the definition. Takes form of + * array('element' => array('attr', 'attr2'), 'element2') + */ + public $needed = array(); + + /** + * Index of inputTokens to rewind to. + */ + protected $rewind = false; + + /** + * Rewind to a spot to re-perform processing. This is useful if you + * deleted a node, and now need to see if this change affected any + * earlier nodes. Rewinding does not affect other injectors, and can + * result in infinite loops if not used carefully. + * @warning HTML Purifier will prevent you from fast-forwarding with this + * function. + */ + public function rewind($index) { + $this->rewind = $index; + } + + /** + * Retrieves rewind, and then unsets it. + */ + public function getRewind() { + $r = $this->rewind; + $this->rewind = false; + return $r; + } + + /** + * Prepares the injector by giving it the config and context objects: + * this allows references to important variables to be made within + * the injector. This function also checks if the HTML environment + * will work with the Injector (see checkNeeded()). + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return Boolean false if success, string of missing needed element/attribute if failure + */ + public function prepare($config, $context) { + $this->htmlDefinition = $config->getHTMLDefinition(); + // Even though this might fail, some unit tests ignore this and + // still test checkNeeded, so be careful. Maybe get rid of that + // dependency. + $result = $this->checkNeeded($config); + if ($result !== false) return $result; + $this->currentNesting =& $context->get('CurrentNesting'); + $this->inputTokens =& $context->get('InputTokens'); + $this->inputIndex =& $context->get('InputIndex'); + return false; + } + + /** + * This function checks if the HTML environment + * will work with the Injector: if p tags are not allowed, the + * Auto-Paragraphing injector should not be enabled. + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return Boolean false if success, string of missing needed element/attribute if failure + */ + public function checkNeeded($config) { + $def = $config->getHTMLDefinition(); + foreach ($this->needed as $element => $attributes) { + if (is_int($element)) $element = $attributes; + if (!isset($def->info[$element])) return $element; + if (!is_array($attributes)) continue; + foreach ($attributes as $name) { + if (!isset($def->info[$element]->attr[$name])) return "$element.$name"; + } + } + return false; + } + + /** + * Tests if the context node allows a certain element + * @param $name Name of element to test for + * @return True if element is allowed, false if it is not + */ + public function allowsElement($name) { + if (!empty($this->currentNesting)) { + $parent_token = array_pop($this->currentNesting); + $this->currentNesting[] = $parent_token; + $parent = $this->htmlDefinition->info[$parent_token->name]; + } else { + $parent = $this->htmlDefinition->info_parent_def; + } + if (!isset($parent->child->elements[$name]) || isset($parent->excludes[$name])) { + return false; + } + // check for exclusion + for ($i = count($this->currentNesting) - 2; $i >= 0; $i--) { + $node = $this->currentNesting[$i]; + $def = $this->htmlDefinition->info[$node->name]; + if (isset($def->excludes[$name])) return false; + } + return true; + } + + /** + * Iterator function, which starts with the next token and continues until + * you reach the end of the input tokens. + * @warning Please prevent previous references from interfering with this + * functions by setting $i = null beforehand! + * @param &$i Current integer index variable for inputTokens + * @param &$current Current token variable. Do NOT use $token, as that variable is also a reference + */ + protected function forward(&$i, &$current) { + if ($i === null) $i = $this->inputIndex + 1; + else $i++; + if (!isset($this->inputTokens[$i])) return false; + $current = $this->inputTokens[$i]; + return true; + } + + /** + * Similar to _forward, but accepts a third parameter $nesting (which + * should be initialized at 0) and stops when we hit the end tag + * for the node $this->inputIndex starts in. + */ + protected function forwardUntilEndToken(&$i, &$current, &$nesting) { + $result = $this->forward($i, $current); + if (!$result) return false; + if ($nesting === null) $nesting = 0; + if ($current instanceof HTMLPurifier_Token_Start) $nesting++; + elseif ($current instanceof HTMLPurifier_Token_End) { + if ($nesting <= 0) return false; + $nesting--; + } + return true; + } + + /** + * Iterator function, starts with the previous token and continues until + * you reach the beginning of input tokens. + * @warning Please prevent previous references from interfering with this + * functions by setting $i = null beforehand! + * @param &$i Current integer index variable for inputTokens + * @param &$current Current token variable. Do NOT use $token, as that variable is also a reference + */ + protected function backward(&$i, &$current) { + if ($i === null) $i = $this->inputIndex - 1; + else $i--; + if ($i < 0) return false; + $current = $this->inputTokens[$i]; + return true; + } + + /** + * Initializes the iterator at the current position. Use in a do {} while; + * loop to force the _forward and _backward functions to start at the + * current location. + * @warning Please prevent previous references from interfering with this + * functions by setting $i = null beforehand! + * @param &$i Current integer index variable for inputTokens + * @param &$current Current token variable. Do NOT use $token, as that variable is also a reference + */ + protected function current(&$i, &$current) { + if ($i === null) $i = $this->inputIndex; + $current = $this->inputTokens[$i]; + } + + /** + * Handler that is called when a text token is processed + */ + public function handleText(&$token) {} + + /** + * Handler that is called when a start or empty token is processed + */ + public function handleElement(&$token) {} + + /** + * Handler that is called when an end token is processed + */ + public function handleEnd(&$token) { + $this->notifyEnd($token); + } + + /** + * Notifier that is called when an end token is processed + * @note This differs from handlers in that the token is read-only + * @deprecated + */ + public function notifyEnd($token) {} + + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/AutoParagraph.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/AutoParagraph.php new file mode 100644 index 000000000..afa760892 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/AutoParagraph.php @@ -0,0 +1,345 @@ +<?php + +/** + * Injector that auto paragraphs text in the root node based on + * double-spacing. + * @todo Ensure all states are unit tested, including variations as well. + * @todo Make a graph of the flow control for this Injector. + */ +class HTMLPurifier_Injector_AutoParagraph extends HTMLPurifier_Injector +{ + + public $name = 'AutoParagraph'; + public $needed = array('p'); + + private function _pStart() { + $par = new HTMLPurifier_Token_Start('p'); + $par->armor['MakeWellFormed_TagClosedError'] = true; + return $par; + } + + public function handleText(&$token) { + $text = $token->data; + // Does the current parent allow <p> tags? + if ($this->allowsElement('p')) { + if (empty($this->currentNesting) || strpos($text, "\n\n") !== false) { + // Note that we have differing behavior when dealing with text + // in the anonymous root node, or a node inside the document. + // If the text as a double-newline, the treatment is the same; + // if it doesn't, see the next if-block if you're in the document. + + $i = $nesting = null; + if (!$this->forwardUntilEndToken($i, $current, $nesting) && $token->is_whitespace) { + // State 1.1: ... ^ (whitespace, then document end) + // ---- + // This is a degenerate case + } else { + if (!$token->is_whitespace || $this->_isInline($current)) { + // State 1.2: PAR1 + // ---- + + // State 1.3: PAR1\n\nPAR2 + // ------------ + + // State 1.4: <div>PAR1\n\nPAR2 (see State 2) + // ------------ + $token = array($this->_pStart()); + $this->_splitText($text, $token); + } else { + // State 1.5: \n<hr /> + // -- + } + } + } else { + // State 2: <div>PAR1... (similar to 1.4) + // ---- + + // We're in an element that allows paragraph tags, but we're not + // sure if we're going to need them. + if ($this->_pLookAhead()) { + // State 2.1: <div>PAR1<b>PAR1\n\nPAR2 + // ---- + // Note: This will always be the first child, since any + // previous inline element would have triggered this very + // same routine, and found the double newline. One possible + // exception would be a comment. + $token = array($this->_pStart(), $token); + } else { + // State 2.2.1: <div>PAR1<div> + // ---- + + // State 2.2.2: <div>PAR1<b>PAR1</b></div> + // ---- + } + } + // Is the current parent a <p> tag? + } elseif ( + !empty($this->currentNesting) && + $this->currentNesting[count($this->currentNesting)-1]->name == 'p' + ) { + // State 3.1: ...<p>PAR1 + // ---- + + // State 3.2: ...<p>PAR1\n\nPAR2 + // ------------ + $token = array(); + $this->_splitText($text, $token); + // Abort! + } else { + // State 4.1: ...<b>PAR1 + // ---- + + // State 4.2: ...<b>PAR1\n\nPAR2 + // ------------ + } + } + + public function handleElement(&$token) { + // We don't have to check if we're already in a <p> tag for block + // tokens, because the tag would have been autoclosed by MakeWellFormed. + if ($this->allowsElement('p')) { + if (!empty($this->currentNesting)) { + if ($this->_isInline($token)) { + // State 1: <div>...<b> + // --- + + // Check if this token is adjacent to the parent token + // (seek backwards until token isn't whitespace) + $i = null; + $this->backward($i, $prev); + + if (!$prev instanceof HTMLPurifier_Token_Start) { + // Token wasn't adjacent + + if ( + $prev instanceof HTMLPurifier_Token_Text && + substr($prev->data, -2) === "\n\n" + ) { + // State 1.1.4: <div><p>PAR1</p>\n\n<b> + // --- + + // Quite frankly, this should be handled by splitText + $token = array($this->_pStart(), $token); + } else { + // State 1.1.1: <div><p>PAR1</p><b> + // --- + + // State 1.1.2: <div><br /><b> + // --- + + // State 1.1.3: <div>PAR<b> + // --- + } + + } else { + // State 1.2.1: <div><b> + // --- + + // Lookahead to see if <p> is needed. + if ($this->_pLookAhead()) { + // State 1.3.1: <div><b>PAR1\n\nPAR2 + // --- + $token = array($this->_pStart(), $token); + } else { + // State 1.3.2: <div><b>PAR1</b></div> + // --- + + // State 1.3.3: <div><b>PAR1</b><div></div>\n\n</div> + // --- + } + } + } else { + // State 2.3: ...<div> + // ----- + } + } else { + if ($this->_isInline($token)) { + // State 3.1: <b> + // --- + // This is where the {p} tag is inserted, not reflected in + // inputTokens yet, however. + $token = array($this->_pStart(), $token); + } else { + // State 3.2: <div> + // ----- + } + + $i = null; + if ($this->backward($i, $prev)) { + if ( + !$prev instanceof HTMLPurifier_Token_Text + ) { + // State 3.1.1: ...</p>{p}<b> + // --- + + // State 3.2.1: ...</p><div> + // ----- + + if (!is_array($token)) $token = array($token); + array_unshift($token, new HTMLPurifier_Token_Text("\n\n")); + } else { + // State 3.1.2: ...</p>\n\n{p}<b> + // --- + + // State 3.2.2: ...</p>\n\n<div> + // ----- + + // Note: PAR<ELEM> cannot occur because PAR would have been + // wrapped in <p> tags. + } + } + } + } else { + // State 2.2: <ul><li> + // ---- + + // State 2.4: <p><b> + // --- + } + } + + /** + * Splits up a text in paragraph tokens and appends them + * to the result stream that will replace the original + * @param $data String text data that will be processed + * into paragraphs + * @param $result Reference to array of tokens that the + * tags will be appended onto + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + */ + private function _splitText($data, &$result) { + $raw_paragraphs = explode("\n\n", $data); + $paragraphs = array(); // without empty paragraphs + $needs_start = false; + $needs_end = false; + + $c = count($raw_paragraphs); + if ($c == 1) { + // There were no double-newlines, abort quickly. In theory this + // should never happen. + $result[] = new HTMLPurifier_Token_Text($data); + return; + } + for ($i = 0; $i < $c; $i++) { + $par = $raw_paragraphs[$i]; + if (trim($par) !== '') { + $paragraphs[] = $par; + } else { + if ($i == 0) { + // Double newline at the front + if (empty($result)) { + // The empty result indicates that the AutoParagraph + // injector did not add any start paragraph tokens. + // This means that we have been in a paragraph for + // a while, and the newline means we should start a new one. + $result[] = new HTMLPurifier_Token_End('p'); + $result[] = new HTMLPurifier_Token_Text("\n\n"); + // However, the start token should only be added if + // there is more processing to be done (i.e. there are + // real paragraphs in here). If there are none, the + // next start paragraph tag will be handled by the + // next call to the injector + $needs_start = true; + } else { + // We just started a new paragraph! + // Reinstate a double-newline for presentation's sake, since + // it was in the source code. + array_unshift($result, new HTMLPurifier_Token_Text("\n\n")); + } + } elseif ($i + 1 == $c) { + // Double newline at the end + // There should be a trailing </p> when we're finally done. + $needs_end = true; + } + } + } + + // Check if this was just a giant blob of whitespace. Move this earlier, + // perhaps? + if (empty($paragraphs)) { + return; + } + + // Add the start tag indicated by \n\n at the beginning of $data + if ($needs_start) { + $result[] = $this->_pStart(); + } + + // Append the paragraphs onto the result + foreach ($paragraphs as $par) { + $result[] = new HTMLPurifier_Token_Text($par); + $result[] = new HTMLPurifier_Token_End('p'); + $result[] = new HTMLPurifier_Token_Text("\n\n"); + $result[] = $this->_pStart(); + } + + // Remove trailing start token; Injector will handle this later if + // it was indeed needed. This prevents from needing to do a lookahead, + // at the cost of a lookbehind later. + array_pop($result); + + // If there is no need for an end tag, remove all of it and let + // MakeWellFormed close it later. + if (!$needs_end) { + array_pop($result); // removes \n\n + array_pop($result); // removes </p> + } + + } + + /** + * Returns true if passed token is inline (and, ergo, allowed in + * paragraph tags) + */ + private function _isInline($token) { + return isset($this->htmlDefinition->info['p']->child->elements[$token->name]); + } + + /** + * Looks ahead in the token list and determines whether or not we need + * to insert a <p> tag. + */ + private function _pLookAhead() { + $this->current($i, $current); + if ($current instanceof HTMLPurifier_Token_Start) $nesting = 1; + else $nesting = 0; + $ok = false; + while ($this->forwardUntilEndToken($i, $current, $nesting)) { + $result = $this->_checkNeedsP($current); + if ($result !== null) { + $ok = $result; + break; + } + } + return $ok; + } + + /** + * Determines if a particular token requires an earlier inline token + * to get a paragraph. This should be used with _forwardUntilEndToken + */ + private function _checkNeedsP($current) { + if ($current instanceof HTMLPurifier_Token_Start){ + if (!$this->_isInline($current)) { + // <div>PAR1<div> + // ---- + // Terminate early, since we hit a block element + return false; + } + } elseif ($current instanceof HTMLPurifier_Token_Text) { + if (strpos($current->data, "\n\n") !== false) { + // <div>PAR1<b>PAR1\n\nPAR2 + // ---- + return true; + } else { + // <div>PAR1<b>PAR1... + // ---- + } + } + return null; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/DisplayLinkURI.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/DisplayLinkURI.php new file mode 100644 index 000000000..9dce9bd08 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/DisplayLinkURI.php @@ -0,0 +1,26 @@ +<?php + +/** + * Injector that displays the URL of an anchor instead of linking to it, in addition to showing the text of the link. + */ +class HTMLPurifier_Injector_DisplayLinkURI extends HTMLPurifier_Injector +{ + + public $name = 'DisplayLinkURI'; + public $needed = array('a'); + + public function handleElement(&$token) { + } + + public function handleEnd(&$token) { + if (isset($token->start->attr['href'])){ + $url = $token->start->attr['href']; + unset($token->start->attr['href']); + $token = array($token, new HTMLPurifier_Token_Text(" ($url)")); + } else { + // nothing to display + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/Linkify.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/Linkify.php new file mode 100644 index 000000000..296dac282 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/Linkify.php @@ -0,0 +1,46 @@ +<?php + +/** + * Injector that converts http, https and ftp text URLs to actual links. + */ +class HTMLPurifier_Injector_Linkify extends HTMLPurifier_Injector +{ + + public $name = 'Linkify'; + public $needed = array('a' => array('href')); + + public function handleText(&$token) { + if (!$this->allowsElement('a')) return; + + if (strpos($token->data, '://') === false) { + // our really quick heuristic failed, abort + // this may not work so well if we want to match things like + // "google.com", but then again, most people don't + return; + } + + // there is/are URL(s). Let's split the string: + // Note: this regex is extremely permissive + $bits = preg_split('#((?:https?|ftp)://[^\s\'"<>()]+)#S', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE); + + $token = array(); + + // $i = index + // $c = count + // $l = is link + for ($i = 0, $c = count($bits), $l = false; $i < $c; $i++, $l = !$l) { + if (!$l) { + if ($bits[$i] === '') continue; + $token[] = new HTMLPurifier_Token_Text($bits[$i]); + } else { + $token[] = new HTMLPurifier_Token_Start('a', array('href' => $bits[$i])); + $token[] = new HTMLPurifier_Token_Text($bits[$i]); + $token[] = new HTMLPurifier_Token_End('a'); + } + } + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/PurifierLinkify.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/PurifierLinkify.php new file mode 100644 index 000000000..ad2455a91 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/PurifierLinkify.php @@ -0,0 +1,45 @@ +<?php + +/** + * Injector that converts configuration directive syntax %Namespace.Directive + * to links + */ +class HTMLPurifier_Injector_PurifierLinkify extends HTMLPurifier_Injector +{ + + public $name = 'PurifierLinkify'; + public $docURL; + public $needed = array('a' => array('href')); + + public function prepare($config, $context) { + $this->docURL = $config->get('AutoFormat.PurifierLinkify.DocURL'); + return parent::prepare($config, $context); + } + + public function handleText(&$token) { + if (!$this->allowsElement('a')) return; + if (strpos($token->data, '%') === false) return; + + $bits = preg_split('#%([a-z0-9]+\.[a-z0-9]+)#Si', $token->data, -1, PREG_SPLIT_DELIM_CAPTURE); + $token = array(); + + // $i = index + // $c = count + // $l = is link + for ($i = 0, $c = count($bits), $l = false; $i < $c; $i++, $l = !$l) { + if (!$l) { + if ($bits[$i] === '') continue; + $token[] = new HTMLPurifier_Token_Text($bits[$i]); + } else { + $token[] = new HTMLPurifier_Token_Start('a', + array('href' => str_replace('%s', $bits[$i], $this->docURL))); + $token[] = new HTMLPurifier_Token_Text('%' . $bits[$i]); + $token[] = new HTMLPurifier_Token_End('a'); + } + } + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveEmpty.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveEmpty.php new file mode 100644 index 000000000..638bfca03 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveEmpty.php @@ -0,0 +1,51 @@ +<?php + +class HTMLPurifier_Injector_RemoveEmpty extends HTMLPurifier_Injector +{ + + private $context, $config, $attrValidator, $removeNbsp, $removeNbspExceptions; + + public function prepare($config, $context) { + parent::prepare($config, $context); + $this->config = $config; + $this->context = $context; + $this->removeNbsp = $config->get('AutoFormat.RemoveEmpty.RemoveNbsp'); + $this->removeNbspExceptions = $config->get('AutoFormat.RemoveEmpty.RemoveNbsp.Exceptions'); + $this->attrValidator = new HTMLPurifier_AttrValidator(); + } + + public function handleElement(&$token) { + if (!$token instanceof HTMLPurifier_Token_Start) return; + $next = false; + for ($i = $this->inputIndex + 1, $c = count($this->inputTokens); $i < $c; $i++) { + $next = $this->inputTokens[$i]; + if ($next instanceof HTMLPurifier_Token_Text) { + if ($next->is_whitespace) continue; + if ($this->removeNbsp && !isset($this->removeNbspExceptions[$token->name])) { + $plain = str_replace("\xC2\xA0", "", $next->data); + $isWsOrNbsp = $plain === '' || ctype_space($plain); + if ($isWsOrNbsp) continue; + } + } + break; + } + if (!$next || ($next instanceof HTMLPurifier_Token_End && $next->name == $token->name)) { + if ($token->name == 'colgroup') return; + $this->attrValidator->validateToken($token, $this->config, $this->context); + $token->armor['ValidateAttributes'] = true; + if (isset($token->attr['id']) || isset($token->attr['name'])) return; + $token = $i - $this->inputIndex + 1; + for ($b = $this->inputIndex - 1; $b > 0; $b--) { + $prev = $this->inputTokens[$b]; + if ($prev instanceof HTMLPurifier_Token_Text && $prev->is_whitespace) continue; + break; + } + // This is safe because we removed the token that triggered this. + $this->rewind($b - 1); + return; + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php new file mode 100644 index 000000000..b21313470 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/RemoveSpansWithoutAttributes.php @@ -0,0 +1,60 @@ +<?php + +/** + * Injector that removes spans with no attributes + */ +class HTMLPurifier_Injector_RemoveSpansWithoutAttributes extends HTMLPurifier_Injector +{ + public $name = 'RemoveSpansWithoutAttributes'; + public $needed = array('span'); + + private $attrValidator; + + /** + * Used by AttrValidator + */ + private $config; + private $context; + + public function prepare($config, $context) { + $this->attrValidator = new HTMLPurifier_AttrValidator(); + $this->config = $config; + $this->context = $context; + return parent::prepare($config, $context); + } + + public function handleElement(&$token) { + if ($token->name !== 'span' || !$token instanceof HTMLPurifier_Token_Start) { + return; + } + + // We need to validate the attributes now since this doesn't normally + // happen until after MakeWellFormed. If all the attributes are removed + // the span needs to be removed too. + $this->attrValidator->validateToken($token, $this->config, $this->context); + $token->armor['ValidateAttributes'] = true; + + if (!empty($token->attr)) { + return; + } + + $nesting = 0; + $spanContentTokens = array(); + while ($this->forwardUntilEndToken($i, $current, $nesting)) {} + + if ($current instanceof HTMLPurifier_Token_End && $current->name === 'span') { + // Mark closing span tag for deletion + $current->markForDeletion = true; + // Delete open span tag + $token = false; + } + } + + public function handleEnd(&$token) { + if ($token->markForDeletion) { + $token = false; + } + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Injector/SafeObject.php b/lib/htmlpurifier/library/HTMLPurifier/Injector/SafeObject.php new file mode 100644 index 000000000..c1d8b0412 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Injector/SafeObject.php @@ -0,0 +1,91 @@ +<?php + +/** + * Adds important param elements to inside of object in order to make + * things safe. + */ +class HTMLPurifier_Injector_SafeObject extends HTMLPurifier_Injector +{ + public $name = 'SafeObject'; + public $needed = array('object', 'param'); + + protected $objectStack = array(); + protected $paramStack = array(); + + // Keep this synchronized with AttrTransform/SafeParam.php + protected $addParam = array( + 'allowScriptAccess' => 'never', + 'allowNetworking' => 'internal', + ); + protected $allowedParam = array( + 'wmode' => true, + 'movie' => true, + 'flashvars' => true, + 'src' => true, + 'allowFullScreen' => true, // if omitted, assume to be 'false' + ); + + public function prepare($config, $context) { + parent::prepare($config, $context); + } + + public function handleElement(&$token) { + if ($token->name == 'object') { + $this->objectStack[] = $token; + $this->paramStack[] = array(); + $new = array($token); + foreach ($this->addParam as $name => $value) { + $new[] = new HTMLPurifier_Token_Empty('param', array('name' => $name, 'value' => $value)); + } + $token = $new; + } elseif ($token->name == 'param') { + $nest = count($this->currentNesting) - 1; + if ($nest >= 0 && $this->currentNesting[$nest]->name === 'object') { + $i = count($this->objectStack) - 1; + if (!isset($token->attr['name'])) { + $token = false; + return; + } + $n = $token->attr['name']; + // We need this fix because YouTube doesn't supply a data + // attribute, which we need if a type is specified. This is + // *very* Flash specific. + if (!isset($this->objectStack[$i]->attr['data']) && + ($token->attr['name'] == 'movie' || $token->attr['name'] == 'src')) { + $this->objectStack[$i]->attr['data'] = $token->attr['value']; + } + // Check if the parameter is the correct value but has not + // already been added + if ( + !isset($this->paramStack[$i][$n]) && + isset($this->addParam[$n]) && + $token->attr['name'] === $this->addParam[$n] + ) { + // keep token, and add to param stack + $this->paramStack[$i][$n] = true; + } elseif (isset($this->allowedParam[$n])) { + // keep token, don't do anything to it + // (could possibly check for duplicates here) + } else { + $token = false; + } + } else { + // not directly inside an object, DENY! + $token = false; + } + } + } + + public function handleEnd(&$token) { + // This is the WRONG way of handling the object and param stacks; + // we should be inserting them directly on the relevant object tokens + // so that the global stack handling handles it. + if ($token->name == 'object') { + array_pop($this->objectStack); + array_pop($this->paramStack); + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language.php b/lib/htmlpurifier/library/HTMLPurifier/Language.php new file mode 100644 index 000000000..3e2be03b5 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language.php @@ -0,0 +1,163 @@ +<?php + +/** + * Represents a language and defines localizable string formatting and + * other functions, as well as the localized messages for HTML Purifier. + */ +class HTMLPurifier_Language +{ + + /** + * ISO 639 language code of language. Prefers shortest possible version + */ + public $code = 'en'; + + /** + * Fallback language code + */ + public $fallback = false; + + /** + * Array of localizable messages + */ + public $messages = array(); + + /** + * Array of localizable error codes + */ + public $errorNames = array(); + + /** + * True if no message file was found for this language, so English + * is being used instead. Check this if you'd like to notify the + * user that they've used a non-supported language. + */ + public $error = false; + + /** + * Has the language object been loaded yet? + * @todo Make it private, fix usage in HTMLPurifier_LanguageTest + */ + public $_loaded = false; + + /** + * Instances of HTMLPurifier_Config and HTMLPurifier_Context + */ + protected $config, $context; + + public function __construct($config, $context) { + $this->config = $config; + $this->context = $context; + } + + /** + * Loads language object with necessary info from factory cache + * @note This is a lazy loader + */ + public function load() { + if ($this->_loaded) return; + $factory = HTMLPurifier_LanguageFactory::instance(); + $factory->loadLanguage($this->code); + foreach ($factory->keys as $key) { + $this->$key = $factory->cache[$this->code][$key]; + } + $this->_loaded = true; + } + + /** + * Retrieves a localised message. + * @param $key string identifier of message + * @return string localised message + */ + public function getMessage($key) { + if (!$this->_loaded) $this->load(); + if (!isset($this->messages[$key])) return "[$key]"; + return $this->messages[$key]; + } + + /** + * Retrieves a localised error name. + * @param $int integer error number, corresponding to PHP's error + * reporting + * @return string localised message + */ + public function getErrorName($int) { + if (!$this->_loaded) $this->load(); + if (!isset($this->errorNames[$int])) return "[Error: $int]"; + return $this->errorNames[$int]; + } + + /** + * Converts an array list into a string readable representation + */ + public function listify($array) { + $sep = $this->getMessage('Item separator'); + $sep_last = $this->getMessage('Item separator last'); + $ret = ''; + for ($i = 0, $c = count($array); $i < $c; $i++) { + if ($i == 0) { + } elseif ($i + 1 < $c) { + $ret .= $sep; + } else { + $ret .= $sep_last; + } + $ret .= $array[$i]; + } + return $ret; + } + + /** + * Formats a localised message with passed parameters + * @param $key string identifier of message + * @param $args Parameters to substitute in + * @return string localised message + * @todo Implement conditionals? Right now, some messages make + * reference to line numbers, but those aren't always available + */ + public function formatMessage($key, $args = array()) { + if (!$this->_loaded) $this->load(); + if (!isset($this->messages[$key])) return "[$key]"; + $raw = $this->messages[$key]; + $subst = array(); + $generator = false; + foreach ($args as $i => $value) { + if (is_object($value)) { + if ($value instanceof HTMLPurifier_Token) { + // factor this out some time + if (!$generator) $generator = $this->context->get('Generator'); + if (isset($value->name)) $subst['$'.$i.'.Name'] = $value->name; + if (isset($value->data)) $subst['$'.$i.'.Data'] = $value->data; + $subst['$'.$i.'.Compact'] = + $subst['$'.$i.'.Serialized'] = $generator->generateFromToken($value); + // a more complex algorithm for compact representation + // could be introduced for all types of tokens. This + // may need to be factored out into a dedicated class + if (!empty($value->attr)) { + $stripped_token = clone $value; + $stripped_token->attr = array(); + $subst['$'.$i.'.Compact'] = $generator->generateFromToken($stripped_token); + } + $subst['$'.$i.'.Line'] = $value->line ? $value->line : 'unknown'; + } + continue; + } elseif (is_array($value)) { + $keys = array_keys($value); + if (array_keys($keys) === $keys) { + // list + $subst['$'.$i] = $this->listify($value); + } else { + // associative array + // no $i implementation yet, sorry + $subst['$'.$i.'.Keys'] = $this->listify($keys); + $subst['$'.$i.'.Values'] = $this->listify(array_values($value)); + } + continue; + } + $subst['$' . $i] = $value; + } + return strtr($raw, $subst); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/classes/en-x-test.php b/lib/htmlpurifier/library/HTMLPurifier/Language/classes/en-x-test.php new file mode 100644 index 000000000..d52fcb7ac --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/classes/en-x-test.php @@ -0,0 +1,12 @@ +<?php + +// private class for unit testing + +class HTMLPurifier_Language_en_x_test extends HTMLPurifier_Language +{ + + + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-test.php b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-test.php new file mode 100644 index 000000000..1c046f379 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-test.php @@ -0,0 +1,11 @@ +<?php + +// private language message file for unit testing purposes + +$fallback = 'en'; + +$messages = array( + 'HTMLPurifier' => 'HTML Purifier X' +); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-testmini.php b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-testmini.php new file mode 100644 index 000000000..806c83fbf --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en-x-testmini.php @@ -0,0 +1,12 @@ +<?php + +// private language message file for unit testing purposes +// this language file has no class associated with it + +$fallback = 'en'; + +$messages = array( + 'HTMLPurifier' => 'HTML Purifier XNone' +); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en.php b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en.php new file mode 100644 index 000000000..8d7b5736b --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Language/messages/en.php @@ -0,0 +1,63 @@ +<?php + +$fallback = false; + +$messages = array( + +'HTMLPurifier' => 'HTML Purifier', + +// for unit testing purposes +'LanguageFactoryTest: Pizza' => 'Pizza', +'LanguageTest: List' => '$1', +'LanguageTest: Hash' => '$1.Keys; $1.Values', + +'Item separator' => ', ', +'Item separator last' => ' and ', // non-Harvard style + +'ErrorCollector: No errors' => 'No errors detected. However, because error reporting is still incomplete, there may have been errors that the error collector was not notified of; please inspect the output HTML carefully.', +'ErrorCollector: At line' => ' at line $line', +'ErrorCollector: Incidental errors' => 'Incidental errors', + +'Lexer: Unclosed comment' => 'Unclosed comment', +'Lexer: Unescaped lt' => 'Unescaped less-than sign (<) should be <', +'Lexer: Missing gt' => 'Missing greater-than sign (>), previous less-than sign (<) should be escaped', +'Lexer: Missing attribute key' => 'Attribute declaration has no key', +'Lexer: Missing end quote' => 'Attribute declaration has no end quote', +'Lexer: Extracted body' => 'Removed document metadata tags', + +'Strategy_RemoveForeignElements: Tag transform' => '<$1> element transformed into $CurrentToken.Serialized', +'Strategy_RemoveForeignElements: Missing required attribute' => '$CurrentToken.Compact element missing required attribute $1', +'Strategy_RemoveForeignElements: Foreign element to text' => 'Unrecognized $CurrentToken.Serialized tag converted to text', +'Strategy_RemoveForeignElements: Foreign element removed' => 'Unrecognized $CurrentToken.Serialized tag removed', +'Strategy_RemoveForeignElements: Comment removed' => 'Comment containing "$CurrentToken.Data" removed', +'Strategy_RemoveForeignElements: Foreign meta element removed' => 'Unrecognized $CurrentToken.Serialized meta tag and all descendants removed', +'Strategy_RemoveForeignElements: Token removed to end' => 'Tags and text starting from $1 element where removed to end', +'Strategy_RemoveForeignElements: Trailing hyphen in comment removed' => 'Trailing hyphen(s) in comment removed', +'Strategy_RemoveForeignElements: Hyphens in comment collapsed' => 'Double hyphens in comments are not allowed, and were collapsed into single hyphens', + +'Strategy_MakeWellFormed: Unnecessary end tag removed' => 'Unnecessary $CurrentToken.Serialized tag removed', +'Strategy_MakeWellFormed: Unnecessary end tag to text' => 'Unnecessary $CurrentToken.Serialized tag converted to text', +'Strategy_MakeWellFormed: Tag auto closed' => '$1.Compact started on line $1.Line auto-closed by $CurrentToken.Compact', +'Strategy_MakeWellFormed: Tag carryover' => '$1.Compact started on line $1.Line auto-continued into $CurrentToken.Compact', +'Strategy_MakeWellFormed: Stray end tag removed' => 'Stray $CurrentToken.Serialized tag removed', +'Strategy_MakeWellFormed: Stray end tag to text' => 'Stray $CurrentToken.Serialized tag converted to text', +'Strategy_MakeWellFormed: Tag closed by element end' => '$1.Compact tag started on line $1.Line closed by end of $CurrentToken.Serialized', +'Strategy_MakeWellFormed: Tag closed by document end' => '$1.Compact tag started on line $1.Line closed by end of document', + +'Strategy_FixNesting: Node removed' => '$CurrentToken.Compact node removed', +'Strategy_FixNesting: Node excluded' => '$CurrentToken.Compact node removed due to descendant exclusion by ancestor element', +'Strategy_FixNesting: Node reorganized' => 'Contents of $CurrentToken.Compact node reorganized to enforce its content model', +'Strategy_FixNesting: Node contents removed' => 'Contents of $CurrentToken.Compact node removed', + +'AttrValidator: Attributes transformed' => 'Attributes on $CurrentToken.Compact transformed from $1.Keys to $2.Keys', +'AttrValidator: Attribute removed' => '$CurrentAttr.Name attribute on $CurrentToken.Compact removed', + +); + +$errorNames = array( + E_ERROR => 'Error', + E_WARNING => 'Warning', + E_NOTICE => 'Notice' +); + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/LanguageFactory.php b/lib/htmlpurifier/library/HTMLPurifier/LanguageFactory.php new file mode 100644 index 000000000..134ef8c74 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/LanguageFactory.php @@ -0,0 +1,198 @@ +<?php + +/** + * Class responsible for generating HTMLPurifier_Language objects, managing + * caching and fallbacks. + * @note Thanks to MediaWiki for the general logic, although this version + * has been entirely rewritten + * @todo Serialized cache for languages + */ +class HTMLPurifier_LanguageFactory +{ + + /** + * Cache of language code information used to load HTMLPurifier_Language objects + * Structure is: $factory->cache[$language_code][$key] = $value + * @value array map + */ + public $cache; + + /** + * Valid keys in the HTMLPurifier_Language object. Designates which + * variables to slurp out of a message file. + * @value array list + */ + public $keys = array('fallback', 'messages', 'errorNames'); + + /** + * Instance of HTMLPurifier_AttrDef_Lang to validate language codes + * @value object HTMLPurifier_AttrDef_Lang + */ + protected $validator; + + /** + * Cached copy of dirname(__FILE__), directory of current file without + * trailing slash + * @value string filename + */ + protected $dir; + + /** + * Keys whose contents are a hash map and can be merged + * @value array lookup + */ + protected $mergeable_keys_map = array('messages' => true, 'errorNames' => true); + + /** + * Keys whose contents are a list and can be merged + * @value array lookup + */ + protected $mergeable_keys_list = array(); + + /** + * Retrieve sole instance of the factory. + * @param $prototype Optional prototype to overload sole instance with, + * or bool true to reset to default factory. + */ + public static function instance($prototype = null) { + static $instance = null; + if ($prototype !== null) { + $instance = $prototype; + } elseif ($instance === null || $prototype == true) { + $instance = new HTMLPurifier_LanguageFactory(); + $instance->setup(); + } + return $instance; + } + + /** + * Sets up the singleton, much like a constructor + * @note Prevents people from getting this outside of the singleton + */ + public function setup() { + $this->validator = new HTMLPurifier_AttrDef_Lang(); + $this->dir = HTMLPURIFIER_PREFIX . '/HTMLPurifier'; + } + + /** + * Creates a language object, handles class fallbacks + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @param $code Code to override configuration with. Private parameter. + */ + public function create($config, $context, $code = false) { + + // validate language code + if ($code === false) { + $code = $this->validator->validate( + $config->get('Core.Language'), $config, $context + ); + } else { + $code = $this->validator->validate($code, $config, $context); + } + if ($code === false) $code = 'en'; // malformed code becomes English + + $pcode = str_replace('-', '_', $code); // make valid PHP classname + static $depth = 0; // recursion protection + + if ($code == 'en') { + $lang = new HTMLPurifier_Language($config, $context); + } else { + $class = 'HTMLPurifier_Language_' . $pcode; + $file = $this->dir . '/Language/classes/' . $code . '.php'; + if (file_exists($file) || class_exists($class, false)) { + $lang = new $class($config, $context); + } else { + // Go fallback + $raw_fallback = $this->getFallbackFor($code); + $fallback = $raw_fallback ? $raw_fallback : 'en'; + $depth++; + $lang = $this->create($config, $context, $fallback); + if (!$raw_fallback) { + $lang->error = true; + } + $depth--; + } + } + + $lang->code = $code; + + return $lang; + + } + + /** + * Returns the fallback language for language + * @note Loads the original language into cache + * @param $code string language code + */ + public function getFallbackFor($code) { + $this->loadLanguage($code); + return $this->cache[$code]['fallback']; + } + + /** + * Loads language into the cache, handles message file and fallbacks + * @param $code string language code + */ + public function loadLanguage($code) { + static $languages_seen = array(); // recursion guard + + // abort if we've already loaded it + if (isset($this->cache[$code])) return; + + // generate filename + $filename = $this->dir . '/Language/messages/' . $code . '.php'; + + // default fallback : may be overwritten by the ensuing include + $fallback = ($code != 'en') ? 'en' : false; + + // load primary localisation + if (!file_exists($filename)) { + // skip the include: will rely solely on fallback + $filename = $this->dir . '/Language/messages/en.php'; + $cache = array(); + } else { + include $filename; + $cache = compact($this->keys); + } + + // load fallback localisation + if (!empty($fallback)) { + + // infinite recursion guard + if (isset($languages_seen[$code])) { + trigger_error('Circular fallback reference in language ' . + $code, E_USER_ERROR); + $fallback = 'en'; + } + $language_seen[$code] = true; + + // load the fallback recursively + $this->loadLanguage($fallback); + $fallback_cache = $this->cache[$fallback]; + + // merge fallback with current language + foreach ( $this->keys as $key ) { + if (isset($cache[$key]) && isset($fallback_cache[$key])) { + if (isset($this->mergeable_keys_map[$key])) { + $cache[$key] = $cache[$key] + $fallback_cache[$key]; + } elseif (isset($this->mergeable_keys_list[$key])) { + $cache[$key] = array_merge( $fallback_cache[$key], $cache[$key] ); + } + } else { + $cache[$key] = $fallback_cache[$key]; + } + } + + } + + // save to cache for later retrieval + $this->cache[$code] = $cache; + + return; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Length.php b/lib/htmlpurifier/library/HTMLPurifier/Length.php new file mode 100644 index 000000000..8d2a46b7d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Length.php @@ -0,0 +1,115 @@ +<?php + +/** + * Represents a measurable length, with a string numeric magnitude + * and a unit. This object is immutable. + */ +class HTMLPurifier_Length +{ + + /** + * String numeric magnitude. + */ + protected $n; + + /** + * String unit. False is permitted if $n = 0. + */ + protected $unit; + + /** + * Whether or not this length is valid. Null if not calculated yet. + */ + protected $isValid; + + /** + * Lookup array of units recognized by CSS 2.1 + */ + protected static $allowedUnits = array( + 'em' => true, 'ex' => true, 'px' => true, 'in' => true, + 'cm' => true, 'mm' => true, 'pt' => true, 'pc' => true + ); + + /** + * @param number $n Magnitude + * @param string $u Unit + */ + public function __construct($n = '0', $u = false) { + $this->n = (string) $n; + $this->unit = $u !== false ? (string) $u : false; + } + + /** + * @param string $s Unit string, like '2em' or '3.4in' + * @warning Does not perform validation. + */ + static public function make($s) { + if ($s instanceof HTMLPurifier_Length) return $s; + $n_length = strspn($s, '1234567890.+-'); + $n = substr($s, 0, $n_length); + $unit = substr($s, $n_length); + if ($unit === '') $unit = false; + return new HTMLPurifier_Length($n, $unit); + } + + /** + * Validates the number and unit. + */ + protected function validate() { + // Special case: + if ($this->n === '+0' || $this->n === '-0') $this->n = '0'; + if ($this->n === '0' && $this->unit === false) return true; + if (!ctype_lower($this->unit)) $this->unit = strtolower($this->unit); + if (!isset(HTMLPurifier_Length::$allowedUnits[$this->unit])) return false; + // Hack: + $def = new HTMLPurifier_AttrDef_CSS_Number(); + $result = $def->validate($this->n, false, false); + if ($result === false) return false; + $this->n = $result; + return true; + } + + /** + * Returns string representation of number. + */ + public function toString() { + if (!$this->isValid()) return false; + return $this->n . $this->unit; + } + + /** + * Retrieves string numeric magnitude. + */ + public function getN() {return $this->n;} + + /** + * Retrieves string unit. + */ + public function getUnit() {return $this->unit;} + + /** + * Returns true if this length unit is valid. + */ + public function isValid() { + if ($this->isValid === null) $this->isValid = $this->validate(); + return $this->isValid; + } + + /** + * Compares two lengths, and returns 1 if greater, -1 if less and 0 if equal. + * @warning If both values are too large or small, this calculation will + * not work properly + */ + public function compareTo($l) { + if ($l === false) return false; + if ($l->unit !== $this->unit) { + $converter = new HTMLPurifier_UnitConverter(); + $l = $converter->convert($l, $this->unit); + if ($l === false) return false; + } + return $this->n - $l->n; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Lexer.php b/lib/htmlpurifier/library/HTMLPurifier/Lexer.php new file mode 100644 index 000000000..9bdbbbb25 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Lexer.php @@ -0,0 +1,326 @@ +<?php + +/** + * Forgivingly lexes HTML (SGML-style) markup into tokens. + * + * A lexer parses a string of SGML-style markup and converts them into + * corresponding tokens. It doesn't check for well-formedness, although its + * internal mechanism may make this automatic (such as the case of + * HTMLPurifier_Lexer_DOMLex). There are several implementations to choose + * from. + * + * A lexer is HTML-oriented: it might work with XML, but it's not + * recommended, as we adhere to a subset of the specification for optimization + * reasons. This might change in the future. Also, most tokenizers are not + * expected to handle DTDs or PIs. + * + * This class should not be directly instantiated, but you may use create() to + * retrieve a default copy of the lexer. Being a supertype, this class + * does not actually define any implementation, but offers commonly used + * convenience functions for subclasses. + * + * @note The unit tests will instantiate this class for testing purposes, as + * many of the utility functions require a class to be instantiated. + * This means that, even though this class is not runnable, it will + * not be declared abstract. + * + * @par + * + * @note + * We use tokens rather than create a DOM representation because DOM would: + * + * @par + * -# Require more processing and memory to create, + * -# Is not streamable, and + * -# Has the entire document structure (html and body not needed). + * + * @par + * However, DOM is helpful in that it makes it easy to move around nodes + * without a lot of lookaheads to see when a tag is closed. This is a + * limitation of the token system and some workarounds would be nice. + */ +class HTMLPurifier_Lexer +{ + + /** + * Whether or not this lexer implements line-number/column-number tracking. + * If it does, set to true. + */ + public $tracksLineNumbers = false; + + // -- STATIC ---------------------------------------------------------- + + /** + * Retrieves or sets the default Lexer as a Prototype Factory. + * + * By default HTMLPurifier_Lexer_DOMLex will be returned. There are + * a few exceptions involving special features that only DirectLex + * implements. + * + * @note The behavior of this class has changed, rather than accepting + * a prototype object, it now accepts a configuration object. + * To specify your own prototype, set %Core.LexerImpl to it. + * This change in behavior de-singletonizes the lexer object. + * + * @param $config Instance of HTMLPurifier_Config + * @return Concrete lexer. + */ + public static function create($config) { + + if (!($config instanceof HTMLPurifier_Config)) { + $lexer = $config; + trigger_error("Passing a prototype to + HTMLPurifier_Lexer::create() is deprecated, please instead + use %Core.LexerImpl", E_USER_WARNING); + } else { + $lexer = $config->get('Core.LexerImpl'); + } + + $needs_tracking = + $config->get('Core.MaintainLineNumbers') || + $config->get('Core.CollectErrors'); + + $inst = null; + if (is_object($lexer)) { + $inst = $lexer; + } else { + + if (is_null($lexer)) { do { + // auto-detection algorithm + + if ($needs_tracking) { + $lexer = 'DirectLex'; + break; + } + + if ( + class_exists('DOMDocument') && + method_exists('DOMDocument', 'loadHTML') && + !extension_loaded('domxml') + ) { + // check for DOM support, because while it's part of the + // core, it can be disabled compile time. Also, the PECL + // domxml extension overrides the default DOM, and is evil + // and nasty and we shan't bother to support it + $lexer = 'DOMLex'; + } else { + $lexer = 'DirectLex'; + } + + } while(0); } // do..while so we can break + + // instantiate recognized string names + switch ($lexer) { + case 'DOMLex': + $inst = new HTMLPurifier_Lexer_DOMLex(); + break; + case 'DirectLex': + $inst = new HTMLPurifier_Lexer_DirectLex(); + break; + case 'PH5P': + $inst = new HTMLPurifier_Lexer_PH5P(); + break; + default: + throw new HTMLPurifier_Exception("Cannot instantiate unrecognized Lexer type " . htmlspecialchars($lexer)); + } + } + + if (!$inst) throw new HTMLPurifier_Exception('No lexer was instantiated'); + + // once PHP DOM implements native line numbers, or we + // hack out something using XSLT, remove this stipulation + if ($needs_tracking && !$inst->tracksLineNumbers) { + throw new HTMLPurifier_Exception('Cannot use lexer that does not support line numbers with Core.MaintainLineNumbers or Core.CollectErrors (use DirectLex instead)'); + } + + return $inst; + + } + + // -- CONVENIENCE MEMBERS --------------------------------------------- + + public function __construct() { + $this->_entity_parser = new HTMLPurifier_EntityParser(); + } + + /** + * Most common entity to raw value conversion table for special entities. + */ + protected $_special_entity2str = + array( + '"' => '"', + '&' => '&', + '<' => '<', + '>' => '>', + ''' => "'", + ''' => "'", + ''' => "'" + ); + + /** + * Parses special entities into the proper characters. + * + * This string will translate escaped versions of the special characters + * into the correct ones. + * + * @warning + * You should be able to treat the output of this function as + * completely parsed, but that's only because all other entities should + * have been handled previously in substituteNonSpecialEntities() + * + * @param $string String character data to be parsed. + * @returns Parsed character data. + */ + public function parseData($string) { + + // following functions require at least one character + if ($string === '') return ''; + + // subtracts amps that cannot possibly be escaped + $num_amp = substr_count($string, '&') - substr_count($string, '& ') - + ($string[strlen($string)-1] === '&' ? 1 : 0); + + if (!$num_amp) return $string; // abort if no entities + $num_esc_amp = substr_count($string, '&'); + $string = strtr($string, $this->_special_entity2str); + + // code duplication for sake of optimization, see above + $num_amp_2 = substr_count($string, '&') - substr_count($string, '& ') - + ($string[strlen($string)-1] === '&' ? 1 : 0); + + if ($num_amp_2 <= $num_esc_amp) return $string; + + // hmm... now we have some uncommon entities. Use the callback. + $string = $this->_entity_parser->substituteSpecialEntities($string); + return $string; + } + + /** + * Lexes an HTML string into tokens. + * + * @param $string String HTML. + * @return HTMLPurifier_Token array representation of HTML. + */ + public function tokenizeHTML($string, $config, $context) { + trigger_error('Call to abstract class', E_USER_ERROR); + } + + /** + * Translates CDATA sections into regular sections (through escaping). + * + * @param $string HTML string to process. + * @returns HTML with CDATA sections escaped. + */ + protected static function escapeCDATA($string) { + return preg_replace_callback( + '/<!\[CDATA\[(.+?)\]\]>/s', + array('HTMLPurifier_Lexer', 'CDATACallback'), + $string + ); + } + + /** + * Special CDATA case that is especially convoluted for <script> + */ + protected static function escapeCommentedCDATA($string) { + return preg_replace_callback( + '#<!--//--><!\[CDATA\[//><!--(.+?)//--><!\]\]>#s', + array('HTMLPurifier_Lexer', 'CDATACallback'), + $string + ); + } + + /** + * Special Internet Explorer conditional comments should be removed. + */ + protected static function removeIEConditional($string) { + return preg_replace( + '#<!--\[if [^>]+\]>.*?<!\[endif\]-->#si', // probably should generalize for all strings + '', + $string + ); + } + + /** + * Callback function for escapeCDATA() that does the work. + * + * @warning Though this is public in order to let the callback happen, + * calling it directly is not recommended. + * @params $matches PCRE matches array, with index 0 the entire match + * and 1 the inside of the CDATA section. + * @returns Escaped internals of the CDATA section. + */ + protected static function CDATACallback($matches) { + // not exactly sure why the character set is needed, but whatever + return htmlspecialchars($matches[1], ENT_COMPAT, 'UTF-8'); + } + + /** + * Takes a piece of HTML and normalizes it by converting entities, fixing + * encoding, extracting bits, and other good stuff. + * @todo Consider making protected + */ + public function normalize($html, $config, $context) { + + // normalize newlines to \n + if ($config->get('Core.NormalizeNewlines')) { + $html = str_replace("\r\n", "\n", $html); + $html = str_replace("\r", "\n", $html); + } + + if ($config->get('HTML.Trusted')) { + // escape convoluted CDATA + $html = $this->escapeCommentedCDATA($html); + } + + // escape CDATA + $html = $this->escapeCDATA($html); + + $html = $this->removeIEConditional($html); + + // extract body from document if applicable + if ($config->get('Core.ConvertDocumentToFragment')) { + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + $new_html = $this->extractBody($html); + if ($e && $new_html != $html) { + $e->send(E_WARNING, 'Lexer: Extracted body'); + } + $html = $new_html; + } + + // expand entities that aren't the big five + $html = $this->_entity_parser->substituteNonSpecialEntities($html); + + // clean into wellformed UTF-8 string for an SGML context: this has + // to be done after entity expansion because the entities sometimes + // represent non-SGML characters (horror, horror!) + $html = HTMLPurifier_Encoder::cleanUTF8($html); + + // if processing instructions are to removed, remove them now + if ($config->get('Core.RemoveProcessingInstructions')) { + $html = preg_replace('#<\?.+?\?>#s', '', $html); + } + + return $html; + } + + /** + * Takes a string of HTML (fragment or document) and returns the content + * @todo Consider making protected + */ + public function extractBody($html) { + $matches = array(); + $result = preg_match('!<body[^>]*>(.*)</body>!is', $html, $matches); + if ($result) { + return $matches[1]; + } else { + return $html; + } + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Lexer/DOMLex.php b/lib/htmlpurifier/library/HTMLPurifier/Lexer/DOMLex.php new file mode 100644 index 000000000..82f377450 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Lexer/DOMLex.php @@ -0,0 +1,243 @@ +<?php + +/** + * Parser that uses PHP 5's DOM extension (part of the core). + * + * In PHP 5, the DOM XML extension was revamped into DOM and added to the core. + * It gives us a forgiving HTML parser, which we use to transform the HTML + * into a DOM, and then into the tokens. It is blazingly fast (for large + * documents, it performs twenty times faster than + * HTMLPurifier_Lexer_DirectLex,and is the default choice for PHP 5. + * + * @note Any empty elements will have empty tokens associated with them, even if + * this is prohibited by the spec. This is cannot be fixed until the spec + * comes into play. + * + * @note PHP's DOM extension does not actually parse any entities, we use + * our own function to do that. + * + * @warning DOM tends to drop whitespace, which may wreak havoc on indenting. + * If this is a huge problem, due to the fact that HTML is hand + * edited and you are unable to get a parser cache that caches the + * the output of HTML Purifier while keeping the original HTML lying + * around, you may want to run Tidy on the resulting output or use + * HTMLPurifier_DirectLex + */ + +class HTMLPurifier_Lexer_DOMLex extends HTMLPurifier_Lexer +{ + + private $factory; + + public function __construct() { + // setup the factory + parent::__construct(); + $this->factory = new HTMLPurifier_TokenFactory(); + } + + public function tokenizeHTML($html, $config, $context) { + + $html = $this->normalize($html, $config, $context); + + // attempt to armor stray angled brackets that cannot possibly + // form tags and thus are probably being used as emoticons + if ($config->get('Core.AggressivelyFixLt')) { + $char = '[^a-z!\/]'; + $comment = "/<!--(.*?)(-->|\z)/is"; + $html = preg_replace_callback($comment, array($this, 'callbackArmorCommentEntities'), $html); + do { + $old = $html; + $html = preg_replace("/<($char)/i", '<\\1', $html); + } while ($html !== $old); + $html = preg_replace_callback($comment, array($this, 'callbackUndoCommentSubst'), $html); // fix comments + } + + // preprocess html, essential for UTF-8 + $html = $this->wrapHTML($html, $config, $context); + + $doc = new DOMDocument(); + $doc->encoding = 'UTF-8'; // theoretically, the above has this covered + + set_error_handler(array($this, 'muteErrorHandler')); + $doc->loadHTML($html); + restore_error_handler(); + + $tokens = array(); + $this->tokenizeDOM( + $doc->getElementsByTagName('html')->item(0)-> // <html> + getElementsByTagName('body')->item(0)-> // <body> + getElementsByTagName('div')->item(0) // <div> + , $tokens); + return $tokens; + } + + /** + * Iterative function that tokenizes a node, putting it into an accumulator. + * To iterate is human, to recurse divine - L. Peter Deutsch + * @param $node DOMNode to be tokenized. + * @param $tokens Array-list of already tokenized tokens. + * @returns Tokens of node appended to previously passed tokens. + */ + protected function tokenizeDOM($node, &$tokens) { + + $level = 0; + $nodes = array($level => array($node)); + $closingNodes = array(); + do { + while (!empty($nodes[$level])) { + $node = array_shift($nodes[$level]); // FIFO + $collect = $level > 0 ? true : false; + $needEndingTag = $this->createStartNode($node, $tokens, $collect); + if ($needEndingTag) { + $closingNodes[$level][] = $node; + } + if ($node->childNodes && $node->childNodes->length) { + $level++; + $nodes[$level] = array(); + foreach ($node->childNodes as $childNode) { + array_push($nodes[$level], $childNode); + } + } + } + $level--; + if ($level && isset($closingNodes[$level])) { + while($node = array_pop($closingNodes[$level])) { + $this->createEndNode($node, $tokens); + } + } + } while ($level > 0); + } + + /** + * @param $node DOMNode to be tokenized. + * @param $tokens Array-list of already tokenized tokens. + * @param $collect Says whether or start and close are collected, set to + * false at first recursion because it's the implicit DIV + * tag you're dealing with. + * @returns bool if the token needs an endtoken + */ + protected function createStartNode($node, &$tokens, $collect) { + // intercept non element nodes. WE MUST catch all of them, + // but we're not getting the character reference nodes because + // those should have been preprocessed + if ($node->nodeType === XML_TEXT_NODE) { + $tokens[] = $this->factory->createText($node->data); + return false; + } elseif ($node->nodeType === XML_CDATA_SECTION_NODE) { + // undo libxml's special treatment of <script> and <style> tags + $last = end($tokens); + $data = $node->data; + // (note $node->tagname is already normalized) + if ($last instanceof HTMLPurifier_Token_Start && ($last->name == 'script' || $last->name == 'style')) { + $new_data = trim($data); + if (substr($new_data, 0, 4) === '<!--') { + $data = substr($new_data, 4); + if (substr($data, -3) === '-->') { + $data = substr($data, 0, -3); + } else { + // Highly suspicious! Not sure what to do... + } + } + } + $tokens[] = $this->factory->createText($this->parseData($data)); + return false; + } elseif ($node->nodeType === XML_COMMENT_NODE) { + // this is code is only invoked for comments in script/style in versions + // of libxml pre-2.6.28 (regular comments, of course, are still + // handled regularly) + $tokens[] = $this->factory->createComment($node->data); + return false; + } elseif ( + // not-well tested: there may be other nodes we have to grab + $node->nodeType !== XML_ELEMENT_NODE + ) { + return false; + } + + $attr = $node->hasAttributes() ? $this->transformAttrToAssoc($node->attributes) : array(); + + // We still have to make sure that the element actually IS empty + if (!$node->childNodes->length) { + if ($collect) { + $tokens[] = $this->factory->createEmpty($node->tagName, $attr); + } + return false; + } else { + if ($collect) { + $tokens[] = $this->factory->createStart( + $tag_name = $node->tagName, // somehow, it get's dropped + $attr + ); + } + return true; + } + } + + protected function createEndNode($node, &$tokens) { + $tokens[] = $this->factory->createEnd($node->tagName); + } + + + /** + * Converts a DOMNamedNodeMap of DOMAttr objects into an assoc array. + * + * @param $attribute_list DOMNamedNodeMap of DOMAttr objects. + * @returns Associative array of attributes. + */ + protected function transformAttrToAssoc($node_map) { + // NamedNodeMap is documented very well, so we're using undocumented + // features, namely, the fact that it implements Iterator and + // has a ->length attribute + if ($node_map->length === 0) return array(); + $array = array(); + foreach ($node_map as $attr) { + $array[$attr->name] = $attr->value; + } + return $array; + } + + /** + * An error handler that mutes all errors + */ + public function muteErrorHandler($errno, $errstr) {} + + /** + * Callback function for undoing escaping of stray angled brackets + * in comments + */ + public function callbackUndoCommentSubst($matches) { + return '<!--' . strtr($matches[1], array('&'=>'&','<'=>'<')) . $matches[2]; + } + + /** + * Callback function that entity-izes ampersands in comments so that + * callbackUndoCommentSubst doesn't clobber them + */ + public function callbackArmorCommentEntities($matches) { + return '<!--' . str_replace('&', '&', $matches[1]) . $matches[2]; + } + + /** + * Wraps an HTML fragment in the necessary HTML + */ + protected function wrapHTML($html, $config, $context) { + $def = $config->getDefinition('HTML'); + $ret = ''; + + if (!empty($def->doctype->dtdPublic) || !empty($def->doctype->dtdSystem)) { + $ret .= '<!DOCTYPE html '; + if (!empty($def->doctype->dtdPublic)) $ret .= 'PUBLIC "' . $def->doctype->dtdPublic . '" '; + if (!empty($def->doctype->dtdSystem)) $ret .= '"' . $def->doctype->dtdSystem . '" '; + $ret .= '>'; + } + + $ret .= '<html><head>'; + $ret .= '<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />'; + // No protection if $html contains a stray </div>! + $ret .= '</head><body><div>'.$html.'</div></body></html>'; + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Lexer/DirectLex.php b/lib/htmlpurifier/library/HTMLPurifier/Lexer/DirectLex.php new file mode 100644 index 000000000..456e6e190 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Lexer/DirectLex.php @@ -0,0 +1,490 @@ +<?php + +/** + * Our in-house implementation of a parser. + * + * A pure PHP parser, DirectLex has absolutely no dependencies, making + * it a reasonably good default for PHP4. Written with efficiency in mind, + * it can be four times faster than HTMLPurifier_Lexer_PEARSax3, although it + * pales in comparison to HTMLPurifier_Lexer_DOMLex. + * + * @todo Reread XML spec and document differences. + */ +class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer +{ + + public $tracksLineNumbers = true; + + /** + * Whitespace characters for str(c)spn. + */ + protected $_whitespace = "\x20\x09\x0D\x0A"; + + /** + * Callback function for script CDATA fudge + * @param $matches, in form of array(opening tag, contents, closing tag) + */ + protected function scriptCallback($matches) { + return $matches[1] . htmlspecialchars($matches[2], ENT_COMPAT, 'UTF-8') . $matches[3]; + } + + public function tokenizeHTML($html, $config, $context) { + + // special normalization for script tags without any armor + // our "armor" heurstic is a < sign any number of whitespaces after + // the first script tag + if ($config->get('HTML.Trusted')) { + $html = preg_replace_callback('#(<script[^>]*>)(\s*[^<].+?)(</script>)#si', + array($this, 'scriptCallback'), $html); + } + + $html = $this->normalize($html, $config, $context); + + $cursor = 0; // our location in the text + $inside_tag = false; // whether or not we're parsing the inside of a tag + $array = array(); // result array + + // This is also treated to mean maintain *column* numbers too + $maintain_line_numbers = $config->get('Core.MaintainLineNumbers'); + + if ($maintain_line_numbers === null) { + // automatically determine line numbering by checking + // if error collection is on + $maintain_line_numbers = $config->get('Core.CollectErrors'); + } + + if ($maintain_line_numbers) { + $current_line = 1; + $current_col = 0; + $length = strlen($html); + } else { + $current_line = false; + $current_col = false; + $length = false; + } + $context->register('CurrentLine', $current_line); + $context->register('CurrentCol', $current_col); + $nl = "\n"; + // how often to manually recalculate. This will ALWAYS be right, + // but it's pretty wasteful. Set to 0 to turn off + $synchronize_interval = $config->get('Core.DirectLexLineNumberSyncInterval'); + + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + + // for testing synchronization + $loops = 0; + + while(++$loops) { + + // $cursor is either at the start of a token, or inside of + // a tag (i.e. there was a < immediately before it), as indicated + // by $inside_tag + + if ($maintain_line_numbers) { + + // $rcursor, however, is always at the start of a token. + $rcursor = $cursor - (int) $inside_tag; + + // Column number is cheap, so we calculate it every round. + // We're interested at the *end* of the newline string, so + // we need to add strlen($nl) == 1 to $nl_pos before subtracting it + // from our "rcursor" position. + $nl_pos = strrpos($html, $nl, $rcursor - $length); + $current_col = $rcursor - (is_bool($nl_pos) ? 0 : $nl_pos + 1); + + // recalculate lines + if ( + $synchronize_interval && // synchronization is on + $cursor > 0 && // cursor is further than zero + $loops % $synchronize_interval === 0 // time to synchronize! + ) { + $current_line = 1 + $this->substrCount($html, $nl, 0, $cursor); + } + + } + + $position_next_lt = strpos($html, '<', $cursor); + $position_next_gt = strpos($html, '>', $cursor); + + // triggers on "<b>asdf</b>" but not "asdf <b></b>" + // special case to set up context + if ($position_next_lt === $cursor) { + $inside_tag = true; + $cursor++; + } + + if (!$inside_tag && $position_next_lt !== false) { + // We are not inside tag and there still is another tag to parse + $token = new + HTMLPurifier_Token_Text( + $this->parseData( + substr( + $html, $cursor, $position_next_lt - $cursor + ) + ) + ); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_lt - $cursor); + } + $array[] = $token; + $cursor = $position_next_lt + 1; + $inside_tag = true; + continue; + } elseif (!$inside_tag) { + // We are not inside tag but there are no more tags + // If we're already at the end, break + if ($cursor === strlen($html)) break; + // Create Text of rest of string + $token = new + HTMLPurifier_Token_Text( + $this->parseData( + substr( + $html, $cursor + ) + ) + ); + if ($maintain_line_numbers) $token->rawPosition($current_line, $current_col); + $array[] = $token; + break; + } elseif ($inside_tag && $position_next_gt !== false) { + // We are in tag and it is well formed + // Grab the internals of the tag + $strlen_segment = $position_next_gt - $cursor; + + if ($strlen_segment < 1) { + // there's nothing to process! + $token = new HTMLPurifier_Token_Text('<'); + $cursor++; + continue; + } + + $segment = substr($html, $cursor, $strlen_segment); + + if ($segment === false) { + // somehow, we attempted to access beyond the end of + // the string, defense-in-depth, reported by Nate Abele + break; + } + + // Check if it's a comment + if ( + substr($segment, 0, 3) === '!--' + ) { + // re-determine segment length, looking for --> + $position_comment_end = strpos($html, '-->', $cursor); + if ($position_comment_end === false) { + // uh oh, we have a comment that extends to + // infinity. Can't be helped: set comment + // end position to end of string + if ($e) $e->send(E_WARNING, 'Lexer: Unclosed comment'); + $position_comment_end = strlen($html); + $end = true; + } else { + $end = false; + } + $strlen_segment = $position_comment_end - $cursor; + $segment = substr($html, $cursor, $strlen_segment); + $token = new + HTMLPurifier_Token_Comment( + substr( + $segment, 3, $strlen_segment - 3 + ) + ); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $strlen_segment); + } + $array[] = $token; + $cursor = $end ? $position_comment_end : $position_comment_end + 3; + $inside_tag = false; + continue; + } + + // Check if it's an end tag + $is_end_tag = (strpos($segment,'/') === 0); + if ($is_end_tag) { + $type = substr($segment, 1); + $token = new HTMLPurifier_Token_End($type); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $inside_tag = false; + $cursor = $position_next_gt + 1; + continue; + } + + // Check leading character is alnum, if not, we may + // have accidently grabbed an emoticon. Translate into + // text and go our merry way + if (!ctype_alpha($segment[0])) { + // XML: $segment[0] !== '_' && $segment[0] !== ':' + if ($e) $e->send(E_NOTICE, 'Lexer: Unescaped lt'); + $token = new HTMLPurifier_Token_Text('<'); + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $inside_tag = false; + continue; + } + + // Check if it is explicitly self closing, if so, remove + // trailing slash. Remember, we could have a tag like <br>, so + // any later token processing scripts must convert improperly + // classified EmptyTags from StartTags. + $is_self_closing = (strrpos($segment,'/') === $strlen_segment-1); + if ($is_self_closing) { + $strlen_segment--; + $segment = substr($segment, 0, $strlen_segment); + } + + // Check if there are any attributes + $position_first_space = strcspn($segment, $this->_whitespace); + + if ($position_first_space >= $strlen_segment) { + if ($is_self_closing) { + $token = new HTMLPurifier_Token_Empty($segment); + } else { + $token = new HTMLPurifier_Token_Start($segment); + } + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $inside_tag = false; + $cursor = $position_next_gt + 1; + continue; + } + + // Grab out all the data + $type = substr($segment, 0, $position_first_space); + $attribute_string = + trim( + substr( + $segment, $position_first_space + ) + ); + if ($attribute_string) { + $attr = $this->parseAttributeString( + $attribute_string + , $config, $context + ); + } else { + $attr = array(); + } + + if ($is_self_closing) { + $token = new HTMLPurifier_Token_Empty($type, $attr); + } else { + $token = new HTMLPurifier_Token_Start($type, $attr); + } + if ($maintain_line_numbers) { + $token->rawPosition($current_line, $current_col); + $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor); + } + $array[] = $token; + $cursor = $position_next_gt + 1; + $inside_tag = false; + continue; + } else { + // inside tag, but there's no ending > sign + if ($e) $e->send(E_WARNING, 'Lexer: Missing gt'); + $token = new + HTMLPurifier_Token_Text( + '<' . + $this->parseData( + substr($html, $cursor) + ) + ); + if ($maintain_line_numbers) $token->rawPosition($current_line, $current_col); + // no cursor scroll? Hmm... + $array[] = $token; + break; + } + break; + } + + $context->destroy('CurrentLine'); + $context->destroy('CurrentCol'); + return $array; + } + + /** + * PHP 5.0.x compatible substr_count that implements offset and length + */ + protected function substrCount($haystack, $needle, $offset, $length) { + static $oldVersion; + if ($oldVersion === null) { + $oldVersion = version_compare(PHP_VERSION, '5.1', '<'); + } + if ($oldVersion) { + $haystack = substr($haystack, $offset, $length); + return substr_count($haystack, $needle); + } else { + return substr_count($haystack, $needle, $offset, $length); + } + } + + /** + * Takes the inside of an HTML tag and makes an assoc array of attributes. + * + * @param $string Inside of tag excluding name. + * @returns Assoc array of attributes. + */ + public function parseAttributeString($string, $config, $context) { + $string = (string) $string; // quick typecast + + if ($string == '') return array(); // no attributes + + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + + // let's see if we can abort as quickly as possible + // one equal sign, no spaces => one attribute + $num_equal = substr_count($string, '='); + $has_space = strpos($string, ' '); + if ($num_equal === 0 && !$has_space) { + // bool attribute + return array($string => $string); + } elseif ($num_equal === 1 && !$has_space) { + // only one attribute + list($key, $quoted_value) = explode('=', $string); + $quoted_value = trim($quoted_value); + if (!$key) { + if ($e) $e->send(E_ERROR, 'Lexer: Missing attribute key'); + return array(); + } + if (!$quoted_value) return array($key => ''); + $first_char = @$quoted_value[0]; + $last_char = @$quoted_value[strlen($quoted_value)-1]; + + $same_quote = ($first_char == $last_char); + $open_quote = ($first_char == '"' || $first_char == "'"); + + if ( $same_quote && $open_quote) { + // well behaved + $value = substr($quoted_value, 1, strlen($quoted_value) - 2); + } else { + // not well behaved + if ($open_quote) { + if ($e) $e->send(E_ERROR, 'Lexer: Missing end quote'); + $value = substr($quoted_value, 1); + } else { + $value = $quoted_value; + } + } + if ($value === false) $value = ''; + return array($key => $this->parseData($value)); + } + + // setup loop environment + $array = array(); // return assoc array of attributes + $cursor = 0; // current position in string (moves forward) + $size = strlen($string); // size of the string (stays the same) + + // if we have unquoted attributes, the parser expects a terminating + // space, so let's guarantee that there's always a terminating space. + $string .= ' '; + + while(true) { + + if ($cursor >= $size) { + break; + } + + $cursor += ($value = strspn($string, $this->_whitespace, $cursor)); + // grab the key + + $key_begin = $cursor; //we're currently at the start of the key + + // scroll past all characters that are the key (not whitespace or =) + $cursor += strcspn($string, $this->_whitespace . '=', $cursor); + + $key_end = $cursor; // now at the end of the key + + $key = substr($string, $key_begin, $key_end - $key_begin); + + if (!$key) { + if ($e) $e->send(E_ERROR, 'Lexer: Missing attribute key'); + $cursor += strcspn($string, $this->_whitespace, $cursor + 1); // prevent infinite loop + continue; // empty key + } + + // scroll past all whitespace + $cursor += strspn($string, $this->_whitespace, $cursor); + + if ($cursor >= $size) { + $array[$key] = $key; + break; + } + + // if the next character is an equal sign, we've got a regular + // pair, otherwise, it's a bool attribute + $first_char = @$string[$cursor]; + + if ($first_char == '=') { + // key="value" + + $cursor++; + $cursor += strspn($string, $this->_whitespace, $cursor); + + if ($cursor === false) { + $array[$key] = ''; + break; + } + + // we might be in front of a quote right now + + $char = @$string[$cursor]; + + if ($char == '"' || $char == "'") { + // it's quoted, end bound is $char + $cursor++; + $value_begin = $cursor; + $cursor = strpos($string, $char, $cursor); + $value_end = $cursor; + } else { + // it's not quoted, end bound is whitespace + $value_begin = $cursor; + $cursor += strcspn($string, $this->_whitespace, $cursor); + $value_end = $cursor; + } + + // we reached a premature end + if ($cursor === false) { + $cursor = $size; + $value_end = $cursor; + } + + $value = substr($string, $value_begin, $value_end - $value_begin); + if ($value === false) $value = ''; + $array[$key] = $this->parseData($value); + $cursor++; + + } else { + // boolattr + if ($key !== '') { + $array[$key] = $key; + } else { + // purely theoretical + if ($e) $e->send(E_ERROR, 'Lexer: Missing attribute key'); + } + + } + } + return $array; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Lexer/PH5P.php b/lib/htmlpurifier/library/HTMLPurifier/Lexer/PH5P.php new file mode 100644 index 000000000..faf00b829 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Lexer/PH5P.php @@ -0,0 +1,3904 @@ +<?php + +/** + * Experimental HTML5-based parser using Jeroen van der Meer's PH5P library. + * Occupies space in the HTML5 pseudo-namespace, which may cause conflicts. + * + * @note + * Recent changes to PHP's DOM extension have resulted in some fatal + * error conditions with the original version of PH5P. Pending changes, + * this lexer will punt to DirectLex if DOM throughs an exception. + */ + +class HTMLPurifier_Lexer_PH5P extends HTMLPurifier_Lexer_DOMLex { + + public function tokenizeHTML($html, $config, $context) { + $new_html = $this->normalize($html, $config, $context); + $new_html = $this->wrapHTML($new_html, $config, $context); + try { + $parser = new HTML5($new_html); + $doc = $parser->save(); + } catch (DOMException $e) { + // Uh oh, it failed. Punt to DirectLex. + $lexer = new HTMLPurifier_Lexer_DirectLex(); + $context->register('PH5PError', $e); // save the error, so we can detect it + return $lexer->tokenizeHTML($html, $config, $context); // use original HTML + } + $tokens = array(); + $this->tokenizeDOM( + $doc->getElementsByTagName('html')->item(0)-> // <html> + getElementsByTagName('body')->item(0)-> // <body> + getElementsByTagName('div')->item(0) // <div> + , $tokens); + return $tokens; + } + +} + +/* + +Copyright 2007 Jeroen van der Meer <http://jero.net/> + +Permission is hereby granted, free of charge, to any person obtaining a +copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to +permit persons to whom the Software is furnished to do so, subject to +the following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. +IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY +CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, +TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE +SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + +*/ + +class HTML5 { + private $data; + private $char; + private $EOF; + private $state; + private $tree; + private $token; + private $content_model; + private $escape = false; + private $entities = array('AElig;','AElig','AMP;','AMP','Aacute;','Aacute', + 'Acirc;','Acirc','Agrave;','Agrave','Alpha;','Aring;','Aring','Atilde;', + 'Atilde','Auml;','Auml','Beta;','COPY;','COPY','Ccedil;','Ccedil','Chi;', + 'Dagger;','Delta;','ETH;','ETH','Eacute;','Eacute','Ecirc;','Ecirc','Egrave;', + 'Egrave','Epsilon;','Eta;','Euml;','Euml','GT;','GT','Gamma;','Iacute;', + 'Iacute','Icirc;','Icirc','Igrave;','Igrave','Iota;','Iuml;','Iuml','Kappa;', + 'LT;','LT','Lambda;','Mu;','Ntilde;','Ntilde','Nu;','OElig;','Oacute;', + 'Oacute','Ocirc;','Ocirc','Ograve;','Ograve','Omega;','Omicron;','Oslash;', + 'Oslash','Otilde;','Otilde','Ouml;','Ouml','Phi;','Pi;','Prime;','Psi;', + 'QUOT;','QUOT','REG;','REG','Rho;','Scaron;','Sigma;','THORN;','THORN', + 'TRADE;','Tau;','Theta;','Uacute;','Uacute','Ucirc;','Ucirc','Ugrave;', + 'Ugrave','Upsilon;','Uuml;','Uuml','Xi;','Yacute;','Yacute','Yuml;','Zeta;', + 'aacute;','aacute','acirc;','acirc','acute;','acute','aelig;','aelig', + 'agrave;','agrave','alefsym;','alpha;','amp;','amp','and;','ang;','apos;', + 'aring;','aring','asymp;','atilde;','atilde','auml;','auml','bdquo;','beta;', + 'brvbar;','brvbar','bull;','cap;','ccedil;','ccedil','cedil;','cedil', + 'cent;','cent','chi;','circ;','clubs;','cong;','copy;','copy','crarr;', + 'cup;','curren;','curren','dArr;','dagger;','darr;','deg;','deg','delta;', + 'diams;','divide;','divide','eacute;','eacute','ecirc;','ecirc','egrave;', + 'egrave','empty;','emsp;','ensp;','epsilon;','equiv;','eta;','eth;','eth', + 'euml;','euml','euro;','exist;','fnof;','forall;','frac12;','frac12', + 'frac14;','frac14','frac34;','frac34','frasl;','gamma;','ge;','gt;','gt', + 'hArr;','harr;','hearts;','hellip;','iacute;','iacute','icirc;','icirc', + 'iexcl;','iexcl','igrave;','igrave','image;','infin;','int;','iota;', + 'iquest;','iquest','isin;','iuml;','iuml','kappa;','lArr;','lambda;','lang;', + 'laquo;','laquo','larr;','lceil;','ldquo;','le;','lfloor;','lowast;','loz;', + 'lrm;','lsaquo;','lsquo;','lt;','lt','macr;','macr','mdash;','micro;','micro', + 'middot;','middot','minus;','mu;','nabla;','nbsp;','nbsp','ndash;','ne;', + 'ni;','not;','not','notin;','nsub;','ntilde;','ntilde','nu;','oacute;', + 'oacute','ocirc;','ocirc','oelig;','ograve;','ograve','oline;','omega;', + 'omicron;','oplus;','or;','ordf;','ordf','ordm;','ordm','oslash;','oslash', + 'otilde;','otilde','otimes;','ouml;','ouml','para;','para','part;','permil;', + 'perp;','phi;','pi;','piv;','plusmn;','plusmn','pound;','pound','prime;', + 'prod;','prop;','psi;','quot;','quot','rArr;','radic;','rang;','raquo;', + 'raquo','rarr;','rceil;','rdquo;','real;','reg;','reg','rfloor;','rho;', + 'rlm;','rsaquo;','rsquo;','sbquo;','scaron;','sdot;','sect;','sect','shy;', + 'shy','sigma;','sigmaf;','sim;','spades;','sub;','sube;','sum;','sup1;', + 'sup1','sup2;','sup2','sup3;','sup3','sup;','supe;','szlig;','szlig','tau;', + 'there4;','theta;','thetasym;','thinsp;','thorn;','thorn','tilde;','times;', + 'times','trade;','uArr;','uacute;','uacute','uarr;','ucirc;','ucirc', + 'ugrave;','ugrave','uml;','uml','upsih;','upsilon;','uuml;','uuml','weierp;', + 'xi;','yacute;','yacute','yen;','yen','yuml;','yuml','zeta;','zwj;','zwnj;'); + + const PCDATA = 0; + const RCDATA = 1; + const CDATA = 2; + const PLAINTEXT = 3; + + const DOCTYPE = 0; + const STARTTAG = 1; + const ENDTAG = 2; + const COMMENT = 3; + const CHARACTR = 4; + const EOF = 5; + + public function __construct($data) { + + $this->data = $data; + $this->char = -1; + $this->EOF = strlen($data); + $this->tree = new HTML5TreeConstructer; + $this->content_model = self::PCDATA; + + $this->state = 'data'; + + while($this->state !== null) { + $this->{$this->state.'State'}(); + } + } + + public function save() { + return $this->tree->save(); + } + + private function char() { + return ($this->char < $this->EOF) + ? $this->data[$this->char] + : false; + } + + private function character($s, $l = 0) { + if($s + $l < $this->EOF) { + if($l === 0) { + return $this->data[$s]; + } else { + return substr($this->data, $s, $l); + } + } + } + + private function characters($char_class, $start) { + return preg_replace('#^(['.$char_class.']+).*#s', '\\1', substr($this->data, $start)); + } + + private function dataState() { + // Consume the next input character + $this->char++; + $char = $this->char(); + + if($char === '&' && ($this->content_model === self::PCDATA || $this->content_model === self::RCDATA)) { + /* U+0026 AMPERSAND (&) + When the content model flag is set to one of the PCDATA or RCDATA + states: switch to the entity data state. Otherwise: treat it as per + the "anything else" entry below. */ + $this->state = 'entityData'; + + } elseif($char === '-') { + /* If the content model flag is set to either the RCDATA state or + the CDATA state, and the escape flag is false, and there are at + least three characters before this one in the input stream, and the + last four characters in the input stream, including this one, are + U+003C LESS-THAN SIGN, U+0021 EXCLAMATION MARK, U+002D HYPHEN-MINUS, + and U+002D HYPHEN-MINUS ("<!--"), then set the escape flag to true. */ + if(($this->content_model === self::RCDATA || $this->content_model === + self::CDATA) && $this->escape === false && + $this->char >= 3 && $this->character($this->char - 4, 4) === '<!--') { + $this->escape = true; + } + + /* In any case, emit the input character as a character token. Stay + in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + /* U+003C LESS-THAN SIGN (<) */ + } elseif($char === '<' && ($this->content_model === self::PCDATA || + (($this->content_model === self::RCDATA || + $this->content_model === self::CDATA) && $this->escape === false))) { + /* When the content model flag is set to the PCDATA state: switch + to the tag open state. + + When the content model flag is set to either the RCDATA state or + the CDATA state and the escape flag is false: switch to the tag + open state. + + Otherwise: treat it as per the "anything else" entry below. */ + $this->state = 'tagOpen'; + + /* U+003E GREATER-THAN SIGN (>) */ + } elseif($char === '>') { + /* If the content model flag is set to either the RCDATA state or + the CDATA state, and the escape flag is true, and the last three + characters in the input stream including this one are U+002D + HYPHEN-MINUS, U+002D HYPHEN-MINUS, U+003E GREATER-THAN SIGN ("-->"), + set the escape flag to false. */ + if(($this->content_model === self::RCDATA || + $this->content_model === self::CDATA) && $this->escape === true && + $this->character($this->char, 3) === '-->') { + $this->escape = false; + } + + /* In any case, emit the input character as a character token. + Stay in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + } elseif($this->char === $this->EOF) { + /* EOF + Emit an end-of-file token. */ + $this->EOF(); + + } elseif($this->content_model === self::PLAINTEXT) { + /* When the content model flag is set to the PLAINTEXT state + THIS DIFFERS GREATLY FROM THE SPEC: Get the remaining characters of + the text and emit it as a character token. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => substr($this->data, $this->char) + )); + + $this->EOF(); + + } else { + /* Anything else + THIS DIFFERS GREATLY FROM THE SPEC: Get as many character that + otherwise would also be treated as a character token and emit it + as a single character token. Stay in the data state. */ + $len = strcspn($this->data, '<&', $this->char); + $char = substr($this->data, $this->char, $len); + $this->char += $len - 1; + + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + $this->state = 'data'; + } + } + + private function entityDataState() { + // Attempt to consume an entity. + $entity = $this->entity(); + + // If nothing is returned, emit a U+0026 AMPERSAND character token. + // Otherwise, emit the character token that was returned. + $char = (!$entity) ? '&' : $entity; + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => $char + )); + + // Finally, switch to the data state. + $this->state = 'data'; + } + + private function tagOpenState() { + switch($this->content_model) { + case self::RCDATA: + case self::CDATA: + /* If the next input character is a U+002F SOLIDUS (/) character, + consume it and switch to the close tag open state. If the next + input character is not a U+002F SOLIDUS (/) character, emit a + U+003C LESS-THAN SIGN character token and switch to the data + state to process the next input character. */ + if($this->character($this->char + 1) === '/') { + $this->char++; + $this->state = 'closeTagOpen'; + + } else { + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '<' + )); + + $this->state = 'data'; + } + break; + + case self::PCDATA: + // If the content model flag is set to the PCDATA state + // Consume the next input character: + $this->char++; + $char = $this->char(); + + if($char === '!') { + /* U+0021 EXCLAMATION MARK (!) + Switch to the markup declaration open state. */ + $this->state = 'markupDeclarationOpen'; + + } elseif($char === '/') { + /* U+002F SOLIDUS (/) + Switch to the close tag open state. */ + $this->state = 'closeTagOpen'; + + } elseif(preg_match('/^[A-Za-z]$/', $char)) { + /* U+0041 LATIN LETTER A through to U+005A LATIN LETTER Z + Create a new start tag token, set its tag name to the lowercase + version of the input character (add 0x0020 to the character's code + point), then switch to the tag name state. (Don't emit the token + yet; further details will be filled in before it is emitted.) */ + $this->token = array( + 'name' => strtolower($char), + 'type' => self::STARTTAG, + 'attr' => array() + ); + + $this->state = 'tagName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Parse error. Emit a U+003C LESS-THAN SIGN character token and a + U+003E GREATER-THAN SIGN character token. Switch to the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '<>' + )); + + $this->state = 'data'; + + } elseif($char === '?') { + /* U+003F QUESTION MARK (?) + Parse error. Switch to the bogus comment state. */ + $this->state = 'bogusComment'; + + } else { + /* Anything else + Parse error. Emit a U+003C LESS-THAN SIGN character token and + reconsume the current input character in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '<' + )); + + $this->char--; + $this->state = 'data'; + } + break; + } + } + + private function closeTagOpenState() { + $next_node = strtolower($this->characters('A-Za-z', $this->char + 1)); + $the_same = count($this->tree->stack) > 0 && $next_node === end($this->tree->stack)->nodeName; + + if(($this->content_model === self::RCDATA || $this->content_model === self::CDATA) && + (!$the_same || ($the_same && (!preg_match('/[\t\n\x0b\x0c >\/]/', + $this->character($this->char + 1 + strlen($next_node))) || $this->EOF === $this->char)))) { + /* If the content model flag is set to the RCDATA or CDATA states then + examine the next few characters. If they do not match the tag name of + the last start tag token emitted (case insensitively), or if they do but + they are not immediately followed by one of the following characters: + * U+0009 CHARACTER TABULATION + * U+000A LINE FEED (LF) + * U+000B LINE TABULATION + * U+000C FORM FEED (FF) + * U+0020 SPACE + * U+003E GREATER-THAN SIGN (>) + * U+002F SOLIDUS (/) + * EOF + ...then there is a parse error. Emit a U+003C LESS-THAN SIGN character + token, a U+002F SOLIDUS character token, and switch to the data state + to process the next input character. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '</' + )); + + $this->state = 'data'; + + } else { + /* Otherwise, if the content model flag is set to the PCDATA state, + or if the next few characters do match that tag name, consume the + next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[A-Za-z]$/', $char)) { + /* U+0041 LATIN LETTER A through to U+005A LATIN LETTER Z + Create a new end tag token, set its tag name to the lowercase version + of the input character (add 0x0020 to the character's code point), then + switch to the tag name state. (Don't emit the token yet; further details + will be filled in before it is emitted.) */ + $this->token = array( + 'name' => strtolower($char), + 'type' => self::ENDTAG + ); + + $this->state = 'tagName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Parse error. Switch to the data state. */ + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit a U+003C LESS-THAN SIGN character token and a U+002F + SOLIDUS character token. Reconsume the EOF character in the data state. */ + $this->emitToken(array( + 'type' => self::CHARACTR, + 'data' => '</' + )); + + $this->char--; + $this->state = 'data'; + + } else { + /* Parse error. Switch to the bogus comment state. */ + $this->state = 'bogusComment'; + } + } + } + + private function tagNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } elseif($char === '/') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Switch to the before + attribute name state. */ + $this->state = 'beforeAttributeName'; + + } else { + /* Anything else + Append the current input character to the current tag token's tag name. + Stay in the tag name state. */ + $this->token['name'] .= strtolower($char); + $this->state = 'tagName'; + } + } + + private function beforeAttributeNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '/') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Stay in the before + attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Start a new attribute in the current tag token. Set that attribute's + name to the current input character, and its value to the empty string. + Switch to the attribute name state. */ + $this->token['attr'][] = array( + 'name' => strtolower($char), + 'value' => null + ); + + $this->state = 'attributeName'; + } + } + + private function attributeNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the before attribute name state. */ + $this->state = 'afterAttributeName'; + + } elseif($char === '=') { + /* U+003D EQUALS SIGN (=) + Switch to the before attribute value state. */ + $this->state = 'beforeAttributeValue'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '/' && $this->character($this->char + 1) !== '>') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Switch to the before + attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's name. + Stay in the attribute name state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['name'] .= strtolower($char); + + $this->state = 'attributeName'; + } + } + + private function afterAttributeNameState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the after attribute name state. */ + $this->state = 'afterAttributeName'; + + } elseif($char === '=') { + /* U+003D EQUALS SIGN (=) + Switch to the before attribute value state. */ + $this->state = 'beforeAttributeValue'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '/' && $this->character($this->char + 1) !== '>') { + /* U+002F SOLIDUS (/) + Parse error unless this is a permitted slash. Switch to the + before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the EOF + character in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Start a new attribute in the current tag token. Set that attribute's + name to the current input character, and its value to the empty string. + Switch to the attribute name state. */ + $this->token['attr'][] = array( + 'name' => strtolower($char), + 'value' => null + ); + + $this->state = 'attributeName'; + } + } + + private function beforeAttributeValueState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Stay in the before attribute value state. */ + $this->state = 'beforeAttributeValue'; + + } elseif($char === '"') { + /* U+0022 QUOTATION MARK (") + Switch to the attribute value (double-quoted) state. */ + $this->state = 'attributeValueDoubleQuoted'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the attribute value (unquoted) state and reconsume + this input character. */ + $this->char--; + $this->state = 'attributeValueUnquoted'; + + } elseif($char === '\'') { + /* U+0027 APOSTROPHE (') + Switch to the attribute value (single-quoted) state. */ + $this->state = 'attributeValueSingleQuoted'; + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Switch to the attribute value (unquoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueUnquoted'; + } + } + + private function attributeValueDoubleQuotedState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if($char === '"') { + /* U+0022 QUOTATION MARK (") + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the entity in attribute value state. */ + $this->entityInAttributeValueState('double'); + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the character + in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Stay in the attribute value (double-quoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueDoubleQuoted'; + } + } + + private function attributeValueSingleQuotedState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if($char === '\'') { + /* U+0022 QUOTATION MARK (') + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the entity in attribute value state. */ + $this->entityInAttributeValueState('single'); + + } elseif($this->char === $this->EOF) { + /* EOF + Parse error. Emit the current tag token. Reconsume the character + in the data state. */ + $this->emitToken($this->token); + + $this->char--; + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Stay in the attribute value (single-quoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueSingleQuoted'; + } + } + + private function attributeValueUnquotedState() { + // Consume the next input character: + $this->char++; + $char = $this->character($this->char); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + /* U+0009 CHARACTER TABULATION + U+000A LINE FEED (LF) + U+000B LINE TABULATION + U+000C FORM FEED (FF) + U+0020 SPACE + Switch to the before attribute name state. */ + $this->state = 'beforeAttributeName'; + + } elseif($char === '&') { + /* U+0026 AMPERSAND (&) + Switch to the entity in attribute value state. */ + $this->entityInAttributeValueState(); + + } elseif($char === '>') { + /* U+003E GREATER-THAN SIGN (>) + Emit the current tag token. Switch to the data state. */ + $this->emitToken($this->token); + $this->state = 'data'; + + } else { + /* Anything else + Append the current input character to the current attribute's value. + Stay in the attribute value (unquoted) state. */ + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + + $this->state = 'attributeValueUnquoted'; + } + } + + private function entityInAttributeValueState() { + // Attempt to consume an entity. + $entity = $this->entity(); + + // If nothing is returned, append a U+0026 AMPERSAND character to the + // current attribute's value. Otherwise, emit the character token that + // was returned. + $char = (!$entity) + ? '&' + : $entity; + + $last = count($this->token['attr']) - 1; + $this->token['attr'][$last]['value'] .= $char; + } + + private function bogusCommentState() { + /* Consume every character up to the first U+003E GREATER-THAN SIGN + character (>) or the end of the file (EOF), whichever comes first. Emit + a comment token whose data is the concatenation of all the characters + starting from and including the character that caused the state machine + to switch into the bogus comment state, up to and including the last + consumed character before the U+003E character, if any, or up to the + end of the file otherwise. (If the comment was started by the end of + the file (EOF), the token is empty.) */ + $data = $this->characters('^>', $this->char); + $this->emitToken(array( + 'data' => $data, + 'type' => self::COMMENT + )); + + $this->char += strlen($data); + + /* Switch to the data state. */ + $this->state = 'data'; + + /* If the end of the file was reached, reconsume the EOF character. */ + if($this->char === $this->EOF) { + $this->char = $this->EOF - 1; + } + } + + private function markupDeclarationOpenState() { + /* If the next two characters are both U+002D HYPHEN-MINUS (-) + characters, consume those two characters, create a comment token whose + data is the empty string, and switch to the comment state. */ + if($this->character($this->char + 1, 2) === '--') { + $this->char += 2; + $this->state = 'comment'; + $this->token = array( + 'data' => null, + 'type' => self::COMMENT + ); + + /* Otherwise if the next seven chacacters are a case-insensitive match + for the word "DOCTYPE", then consume those characters and switch to the + DOCTYPE state. */ + } elseif(strtolower($this->character($this->char + 1, 7)) === 'doctype') { + $this->char += 7; + $this->state = 'doctype'; + + /* Otherwise, is is a parse error. Switch to the bogus comment state. + The next character that is consumed, if any, is the first character + that will be in the comment. */ + } else { + $this->char++; + $this->state = 'bogusComment'; + } + } + + private function commentState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + /* U+002D HYPHEN-MINUS (-) */ + if($char === '-') { + /* Switch to the comment dash state */ + $this->state = 'commentDash'; + + /* EOF */ + } elseif($this->char === $this->EOF) { + /* Parse error. Emit the comment token. Reconsume the EOF character + in the data state. */ + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + /* Anything else */ + } else { + /* Append the input character to the comment token's data. Stay in + the comment state. */ + $this->token['data'] .= $char; + } + } + + private function commentDashState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + /* U+002D HYPHEN-MINUS (-) */ + if($char === '-') { + /* Switch to the comment end state */ + $this->state = 'commentEnd'; + + /* EOF */ + } elseif($this->char === $this->EOF) { + /* Parse error. Emit the comment token. Reconsume the EOF character + in the data state. */ + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + /* Anything else */ + } else { + /* Append a U+002D HYPHEN-MINUS (-) character and the input + character to the comment token's data. Switch to the comment state. */ + $this->token['data'] .= '-'.$char; + $this->state = 'comment'; + } + } + + private function commentEndState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($char === '-') { + $this->token['data'] .= '-'; + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + $this->token['data'] .= '--'.$char; + $this->state = 'comment'; + } + } + + private function doctypeState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + $this->state = 'beforeDoctypeName'; + + } else { + $this->char--; + $this->state = 'beforeDoctypeName'; + } + } + + private function beforeDoctypeNameState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + // Stay in the before DOCTYPE name state. + + } elseif(preg_match('/^[a-z]$/', $char)) { + $this->token = array( + 'name' => strtoupper($char), + 'type' => self::DOCTYPE, + 'error' => true + ); + + $this->state = 'doctypeName'; + + } elseif($char === '>') { + $this->emitToken(array( + 'name' => null, + 'type' => self::DOCTYPE, + 'error' => true + )); + + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + $this->emitToken(array( + 'name' => null, + 'type' => self::DOCTYPE, + 'error' => true + )); + + $this->char--; + $this->state = 'data'; + + } else { + $this->token = array( + 'name' => $char, + 'type' => self::DOCTYPE, + 'error' => true + ); + + $this->state = 'doctypeName'; + } + } + + private function doctypeNameState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + $this->state = 'AfterDoctypeName'; + + } elseif($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif(preg_match('/^[a-z]$/', $char)) { + $this->token['name'] .= strtoupper($char); + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + $this->token['name'] .= $char; + } + + $this->token['error'] = ($this->token['name'] === 'HTML') + ? false + : true; + } + + private function afterDoctypeNameState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if(preg_match('/^[\t\n\x0b\x0c ]$/', $char)) { + // Stay in the DOCTYPE name state. + + } elseif($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + $this->token['error'] = true; + $this->state = 'bogusDoctype'; + } + } + + private function bogusDoctypeState() { + /* Consume the next input character: */ + $this->char++; + $char = $this->char(); + + if($char === '>') { + $this->emitToken($this->token); + $this->state = 'data'; + + } elseif($this->char === $this->EOF) { + $this->emitToken($this->token); + $this->char--; + $this->state = 'data'; + + } else { + // Stay in the bogus DOCTYPE state. + } + } + + private function entity() { + $start = $this->char; + + // This section defines how to consume an entity. This definition is + // used when parsing entities in text and in attributes. + + // The behaviour depends on the identity of the next character (the + // one immediately after the U+0026 AMPERSAND character): + + switch($this->character($this->char + 1)) { + // U+0023 NUMBER SIGN (#) + case '#': + + // The behaviour further depends on the character after the + // U+0023 NUMBER SIGN: + switch($this->character($this->char + 1)) { + // U+0078 LATIN SMALL LETTER X + // U+0058 LATIN CAPITAL LETTER X + case 'x': + case 'X': + // Follow the steps below, but using the range of + // characters U+0030 DIGIT ZERO through to U+0039 DIGIT + // NINE, U+0061 LATIN SMALL LETTER A through to U+0066 + // LATIN SMALL LETTER F, and U+0041 LATIN CAPITAL LETTER + // A, through to U+0046 LATIN CAPITAL LETTER F (in other + // words, 0-9, A-F, a-f). + $char = 1; + $char_class = '0-9A-Fa-f'; + break; + + // Anything else + default: + // Follow the steps below, but using the range of + // characters U+0030 DIGIT ZERO through to U+0039 DIGIT + // NINE (i.e. just 0-9). + $char = 0; + $char_class = '0-9'; + break; + } + + // Consume as many characters as match the range of characters + // given above. + $this->char++; + $e_name = $this->characters($char_class, $this->char + $char + 1); + $entity = $this->character($start, $this->char); + $cond = strlen($e_name) > 0; + + // The rest of the parsing happens bellow. + break; + + // Anything else + default: + // Consume the maximum number of characters possible, with the + // consumed characters case-sensitively matching one of the + // identifiers in the first column of the entities table. + $e_name = $this->characters('0-9A-Za-z;', $this->char + 1); + $len = strlen($e_name); + + for($c = 1; $c <= $len; $c++) { + $id = substr($e_name, 0, $c); + $this->char++; + + if(in_array($id, $this->entities)) { + if ($e_name[$c-1] !== ';') { + if ($c < $len && $e_name[$c] == ';') { + $this->char++; // consume extra semicolon + } + } + $entity = $id; + break; + } + } + + $cond = isset($entity); + // The rest of the parsing happens bellow. + break; + } + + if(!$cond) { + // If no match can be made, then this is a parse error. No + // characters are consumed, and nothing is returned. + $this->char = $start; + return false; + } + + // Return a character token for the character corresponding to the + // entity name (as given by the second column of the entities table). + return html_entity_decode('&'.$entity.';', ENT_QUOTES, 'UTF-8'); + } + + private function emitToken($token) { + $emit = $this->tree->emitToken($token); + + if(is_int($emit)) { + $this->content_model = $emit; + + } elseif($token['type'] === self::ENDTAG) { + $this->content_model = self::PCDATA; + } + } + + private function EOF() { + $this->state = null; + $this->tree->emitToken(array( + 'type' => self::EOF + )); + } +} + +class HTML5TreeConstructer { + public $stack = array(); + + private $phase; + private $mode; + private $dom; + private $foster_parent = null; + private $a_formatting = array(); + + private $head_pointer = null; + private $form_pointer = null; + + private $scoping = array('button','caption','html','marquee','object','table','td','th'); + private $formatting = array('a','b','big','em','font','i','nobr','s','small','strike','strong','tt','u'); + private $special = array('address','area','base','basefont','bgsound', + 'blockquote','body','br','center','col','colgroup','dd','dir','div','dl', + 'dt','embed','fieldset','form','frame','frameset','h1','h2','h3','h4','h5', + 'h6','head','hr','iframe','image','img','input','isindex','li','link', + 'listing','menu','meta','noembed','noframes','noscript','ol','optgroup', + 'option','p','param','plaintext','pre','script','select','spacer','style', + 'tbody','textarea','tfoot','thead','title','tr','ul','wbr'); + + // The different phases. + const INIT_PHASE = 0; + const ROOT_PHASE = 1; + const MAIN_PHASE = 2; + const END_PHASE = 3; + + // The different insertion modes for the main phase. + const BEFOR_HEAD = 0; + const IN_HEAD = 1; + const AFTER_HEAD = 2; + const IN_BODY = 3; + const IN_TABLE = 4; + const IN_CAPTION = 5; + const IN_CGROUP = 6; + const IN_TBODY = 7; + const IN_ROW = 8; + const IN_CELL = 9; + const IN_SELECT = 10; + const AFTER_BODY = 11; + const IN_FRAME = 12; + const AFTR_FRAME = 13; + + // The different types of elements. + const SPECIAL = 0; + const SCOPING = 1; + const FORMATTING = 2; + const PHRASING = 3; + + const MARKER = 0; + + public function __construct() { + $this->phase = self::INIT_PHASE; + $this->mode = self::BEFOR_HEAD; + $this->dom = new DOMDocument; + + $this->dom->encoding = 'UTF-8'; + $this->dom->preserveWhiteSpace = true; + $this->dom->substituteEntities = true; + $this->dom->strictErrorChecking = false; + } + + // Process tag tokens + public function emitToken($token) { + switch($this->phase) { + case self::INIT_PHASE: return $this->initPhase($token); break; + case self::ROOT_PHASE: return $this->rootElementPhase($token); break; + case self::MAIN_PHASE: return $this->mainPhase($token); break; + case self::END_PHASE : return $this->trailingEndPhase($token); break; + } + } + + private function initPhase($token) { + /* Initially, the tree construction stage must handle each token + emitted from the tokenisation stage as follows: */ + + /* A DOCTYPE token that is marked as being in error + A comment token + A start tag token + An end tag token + A character token that is not one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE + An end-of-file token */ + if((isset($token['error']) && $token['error']) || + $token['type'] === HTML5::COMMENT || + $token['type'] === HTML5::STARTTAG || + $token['type'] === HTML5::ENDTAG || + $token['type'] === HTML5::EOF || + ($token['type'] === HTML5::CHARACTR && isset($token['data']) && + !preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data']))) { + /* This specification does not define how to handle this case. In + particular, user agents may ignore the entirety of this specification + altogether for such documents, and instead invoke special parse modes + with a greater emphasis on backwards compatibility. */ + + $this->phase = self::ROOT_PHASE; + return $this->rootElementPhase($token); + + /* A DOCTYPE token marked as being correct */ + } elseif(isset($token['error']) && !$token['error']) { + /* Append a DocumentType node to the Document node, with the name + attribute set to the name given in the DOCTYPE token (which will be + "HTML"), and the other attributes specific to DocumentType objects + set to null, empty lists, or the empty string as appropriate. */ + $doctype = new DOMDocumentType(null, null, 'HTML'); + + /* Then, switch to the root element phase of the tree construction + stage. */ + $this->phase = self::ROOT_PHASE; + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + } elseif(isset($token['data']) && preg_match('/^[\t\n\x0b\x0c ]+$/', + $token['data'])) { + /* Append that character to the Document node. */ + $text = $this->dom->createTextNode($token['data']); + $this->dom->appendChild($text); + } + } + + private function rootElementPhase($token) { + /* After the initial phase, as each token is emitted from the tokenisation + stage, it must be processed as described in this section. */ + + /* A DOCTYPE token */ + if($token['type'] === HTML5::DOCTYPE) { + // Parse error. Ignore the token. + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the Document object with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + $this->dom->appendChild($comment); + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + } elseif($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append that character to the Document node. */ + $text = $this->dom->createTextNode($token['data']); + $this->dom->appendChild($text); + + /* A character token that is not one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED + (FF), or U+0020 SPACE + A start tag token + An end tag token + An end-of-file token */ + } elseif(($token['type'] === HTML5::CHARACTR && + !preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) || + $token['type'] === HTML5::STARTTAG || + $token['type'] === HTML5::ENDTAG || + $token['type'] === HTML5::EOF) { + /* Create an HTMLElement node with the tag name html, in the HTML + namespace. Append it to the Document object. Switch to the main + phase and reprocess the current token. */ + $html = $this->dom->createElement('html'); + $this->dom->appendChild($html); + $this->stack[] = $html; + + $this->phase = self::MAIN_PHASE; + return $this->mainPhase($token); + } + } + + private function mainPhase($token) { + /* Tokens in the main phase must be handled as follows: */ + + /* A DOCTYPE token */ + if($token['type'] === HTML5::DOCTYPE) { + // Parse error. Ignore the token. + + /* A start tag token with the tag name "html" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'html') { + /* If this start tag token was not the first start tag token, then + it is a parse error. */ + + /* For each attribute on the token, check to see if the attribute + is already present on the top element of the stack of open elements. + If it is not, add the attribute and its corresponding value to that + element. */ + foreach($token['attr'] as $attr) { + if(!$this->stack[0]->hasAttribute($attr['name'])) { + $this->stack[0]->setAttribute($attr['name'], $attr['value']); + } + } + + /* An end-of-file token */ + } elseif($token['type'] === HTML5::EOF) { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* Anything else. */ + } else { + /* Depends on the insertion mode: */ + switch($this->mode) { + case self::BEFOR_HEAD: return $this->beforeHead($token); break; + case self::IN_HEAD: return $this->inHead($token); break; + case self::AFTER_HEAD: return $this->afterHead($token); break; + case self::IN_BODY: return $this->inBody($token); break; + case self::IN_TABLE: return $this->inTable($token); break; + case self::IN_CAPTION: return $this->inCaption($token); break; + case self::IN_CGROUP: return $this->inColumnGroup($token); break; + case self::IN_TBODY: return $this->inTableBody($token); break; + case self::IN_ROW: return $this->inRow($token); break; + case self::IN_CELL: return $this->inCell($token); break; + case self::IN_SELECT: return $this->inSelect($token); break; + case self::AFTER_BODY: return $this->afterBody($token); break; + case self::IN_FRAME: return $this->inFrameset($token); break; + case self::AFTR_FRAME: return $this->afterFrameset($token); break; + case self::END_PHASE: return $this->trailingEndPhase($token); break; + } + } + } + + private function beforeHead($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data attribute + set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag token with the tag name "head" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'head') { + /* Create an element for the token, append the new element to the + current node and push it onto the stack of open elements. */ + $element = $this->insertElement($token); + + /* Set the head element pointer to this new element node. */ + $this->head_pointer = $element; + + /* Change the insertion mode to "in head". */ + $this->mode = self::IN_HEAD; + + /* A start tag token whose tag name is one of: "base", "link", "meta", + "script", "style", "title". Or an end tag with the tag name "html". + Or a character token that is not one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE. Or any other start tag token */ + } elseif($token['type'] === HTML5::STARTTAG || + ($token['type'] === HTML5::ENDTAG && $token['name'] === 'html') || + ($token['type'] === HTML5::CHARACTR && !preg_match('/^[\t\n\x0b\x0c ]$/', + $token['data']))) { + /* Act as if a start tag token with the tag name "head" and no + attributes had been seen, then reprocess the current token. */ + $this->beforeHead(array( + 'name' => 'head', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inHead($token); + + /* Any other end tag */ + } elseif($token['type'] === HTML5::ENDTAG) { + /* Parse error. Ignore the token. */ + } + } + + private function inHead($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE. + + THIS DIFFERS FROM THE SPEC: If the current node is either a title, style + or script element, append the character to the current node regardless + of its content. */ + if(($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) || ( + $token['type'] === HTML5::CHARACTR && in_array(end($this->stack)->nodeName, + array('title', 'style', 'script')))) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data attribute + set to the data given in the comment token. */ + $this->insertComment($token['data']); + + } elseif($token['type'] === HTML5::ENDTAG && + in_array($token['name'], array('title', 'style', 'script'))) { + array_pop($this->stack); + return HTML5::PCDATA; + + /* A start tag with the tag name "title" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'title') { + /* Create an element for the token and append the new element to the + node pointed to by the head element pointer, or, if that is null + (innerHTML case), to the current node. */ + if($this->head_pointer !== null) { + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + + } else { + $element = $this->insertElement($token); + } + + /* Switch the tokeniser's content model flag to the RCDATA state. */ + return HTML5::RCDATA; + + /* A start tag with the tag name "style" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'style') { + /* Create an element for the token and append the new element to the + node pointed to by the head element pointer, or, if that is null + (innerHTML case), to the current node. */ + if($this->head_pointer !== null) { + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + + } else { + $this->insertElement($token); + } + + /* Switch the tokeniser's content model flag to the CDATA state. */ + return HTML5::CDATA; + + /* A start tag with the tag name "script" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'script') { + /* Create an element for the token. */ + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + + /* Switch the tokeniser's content model flag to the CDATA state. */ + return HTML5::CDATA; + + /* A start tag with the tag name "base", "link", or "meta" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('base', 'link', 'meta'))) { + /* Create an element for the token and append the new element to the + node pointed to by the head element pointer, or, if that is null + (innerHTML case), to the current node. */ + if($this->head_pointer !== null) { + $element = $this->insertElement($token, false); + $this->head_pointer->appendChild($element); + array_pop($this->stack); + + } else { + $this->insertElement($token); + } + + /* An end tag with the tag name "head" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'head') { + /* If the current node is a head element, pop the current node off + the stack of open elements. */ + if($this->head_pointer->isSameNode(end($this->stack))) { + array_pop($this->stack); + + /* Otherwise, this is a parse error. */ + } else { + // k + } + + /* Change the insertion mode to "after head". */ + $this->mode = self::AFTER_HEAD; + + /* A start tag with the tag name "head" or an end tag except "html". */ + } elseif(($token['type'] === HTML5::STARTTAG && $token['name'] === 'head') || + ($token['type'] === HTML5::ENDTAG && $token['name'] !== 'html')) { + // Parse error. Ignore the token. + + /* Anything else */ + } else { + /* If the current node is a head element, act as if an end tag + token with the tag name "head" had been seen. */ + if($this->head_pointer->isSameNode(end($this->stack))) { + $this->inHead(array( + 'name' => 'head', + 'type' => HTML5::ENDTAG + )); + + /* Otherwise, change the insertion mode to "after head". */ + } else { + $this->mode = self::AFTER_HEAD; + } + + /* Then, reprocess the current token. */ + return $this->afterHead($token); + } + } + + private function afterHead($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data attribute + set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag token with the tag name "body" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'body') { + /* Insert a body element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in body". */ + $this->mode = self::IN_BODY; + + /* A start tag token with the tag name "frameset" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'frameset') { + /* Insert a frameset element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in frameset". */ + $this->mode = self::IN_FRAME; + + /* A start tag token whose tag name is one of: "base", "link", "meta", + "script", "style", "title" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('base', 'link', 'meta', 'script', 'style', 'title'))) { + /* Parse error. Switch the insertion mode back to "in head" and + reprocess the token. */ + $this->mode = self::IN_HEAD; + return $this->inHead($token); + + /* Anything else */ + } else { + /* Act as if a start tag token with the tag name "body" and no + attributes had been seen, and then reprocess the current token. */ + $this->afterHead(array( + 'name' => 'body', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inBody($token); + } + } + + private function inBody($token) { + /* Handle the token as follows: */ + + switch($token['type']) { + /* A character token */ + case HTML5::CHARACTR: + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Append the token's character to the current node. */ + $this->insertText($token['data']); + break; + + /* A comment token */ + case HTML5::COMMENT: + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + break; + + case HTML5::STARTTAG: + switch($token['name']) { + /* A start tag token whose tag name is one of: "script", + "style" */ + case 'script': case 'style': + /* Process the token as if the insertion mode had been "in + head". */ + return $this->inHead($token); + break; + + /* A start tag token whose tag name is one of: "base", "link", + "meta", "title" */ + case 'base': case 'link': case 'meta': case 'title': + /* Parse error. Process the token as if the insertion mode + had been "in head". */ + return $this->inHead($token); + break; + + /* A start tag token with the tag name "body" */ + case 'body': + /* Parse error. If the second element on the stack of open + elements is not a body element, or, if the stack of open + elements has only one node on it, then ignore the token. + (innerHTML case) */ + if(count($this->stack) === 1 || $this->stack[1]->nodeName !== 'body') { + // Ignore + + /* Otherwise, for each attribute on the token, check to see + if the attribute is already present on the body element (the + second element) on the stack of open elements. If it is not, + add the attribute and its corresponding value to that + element. */ + } else { + foreach($token['attr'] as $attr) { + if(!$this->stack[1]->hasAttribute($attr['name'])) { + $this->stack[1]->setAttribute($attr['name'], $attr['value']); + } + } + } + break; + + /* A start tag whose tag name is one of: "address", + "blockquote", "center", "dir", "div", "dl", "fieldset", + "listing", "menu", "ol", "p", "ul" */ + case 'address': case 'blockquote': case 'center': case 'dir': + case 'div': case 'dl': case 'fieldset': case 'listing': + case 'menu': case 'ol': case 'p': case 'ul': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been + seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + break; + + /* A start tag whose tag name is "form" */ + case 'form': + /* If the form element pointer is not null, ignore the + token with a parse error. */ + if($this->form_pointer !== null) { + // Ignore. + + /* Otherwise: */ + } else { + /* If the stack of open elements has a p element in + scope, then act as if an end tag with the tag name p + had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token, and set the + form element pointer to point to the element created. */ + $element = $this->insertElement($token); + $this->form_pointer = $element; + } + break; + + /* A start tag whose tag name is "li", "dd" or "dt" */ + case 'li': case 'dd': case 'dt': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been + seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + $stack_length = count($this->stack) - 1; + + for($n = $stack_length; 0 <= $n; $n--) { + /* 1. Initialise node to be the current node (the + bottommost node of the stack). */ + $stop = false; + $node = $this->stack[$n]; + $cat = $this->getElementCategory($node->tagName); + + /* 2. If node is an li, dd or dt element, then pop all + the nodes from the current node up to node, including + node, then stop this algorithm. */ + if($token['name'] === $node->tagName || ($token['name'] !== 'li' + && ($node->tagName === 'dd' || $node->tagName === 'dt'))) { + for($x = $stack_length; $x >= $n ; $x--) { + array_pop($this->stack); + } + + break; + } + + /* 3. If node is not in the formatting category, and is + not in the phrasing category, and is not an address or + div element, then stop this algorithm. */ + if($cat !== self::FORMATTING && $cat !== self::PHRASING && + $node->tagName !== 'address' && $node->tagName !== 'div') { + break; + } + } + + /* Finally, insert an HTML element with the same tag + name as the token's. */ + $this->insertElement($token); + break; + + /* A start tag token whose tag name is "plaintext" */ + case 'plaintext': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been + seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + return HTML5::PLAINTEXT; + break; + + /* A start tag whose tag name is one of: "h1", "h2", "h3", "h4", + "h5", "h6" */ + case 'h1': case 'h2': case 'h3': case 'h4': case 'h5': case 'h6': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* If the stack of open elements has in scope an element whose + tag name is one of "h1", "h2", "h3", "h4", "h5", or "h6", then + this is a parse error; pop elements from the stack until an + element with one of those tag names has been popped from the + stack. */ + while($this->elementInScope(array('h1', 'h2', 'h3', 'h4', 'h5', 'h6'))) { + array_pop($this->stack); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + break; + + /* A start tag whose tag name is "a" */ + case 'a': + /* If the list of active formatting elements contains + an element whose tag name is "a" between the end of the + list and the last marker on the list (or the start of + the list if there is no marker on the list), then this + is a parse error; act as if an end tag with the tag name + "a" had been seen, then remove that element from the list + of active formatting elements and the stack of open + elements if the end tag didn't already remove it (it + might not have if the element is not in table scope). */ + $leng = count($this->a_formatting); + + for($n = $leng - 1; $n >= 0; $n--) { + if($this->a_formatting[$n] === self::MARKER) { + break; + + } elseif($this->a_formatting[$n]->nodeName === 'a') { + $this->emitToken(array( + 'name' => 'a', + 'type' => HTML5::ENDTAG + )); + break; + } + } + + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $el = $this->insertElement($token); + + /* Add that element to the list of active formatting + elements. */ + $this->a_formatting[] = $el; + break; + + /* A start tag whose tag name is one of: "b", "big", "em", "font", + "i", "nobr", "s", "small", "strike", "strong", "tt", "u" */ + case 'b': case 'big': case 'em': case 'font': case 'i': + case 'nobr': case 's': case 'small': case 'strike': + case 'strong': case 'tt': case 'u': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $el = $this->insertElement($token); + + /* Add that element to the list of active formatting + elements. */ + $this->a_formatting[] = $el; + break; + + /* A start tag token whose tag name is "button" */ + case 'button': + /* If the stack of open elements has a button element in scope, + then this is a parse error; act as if an end tag with the tag + name "button" had been seen, then reprocess the token. (We don't + do that. Unnecessary.) */ + if($this->elementInScope('button')) { + $this->inBody(array( + 'name' => 'button', + 'type' => HTML5::ENDTAG + )); + } + + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Insert a marker at the end of the list of active + formatting elements. */ + $this->a_formatting[] = self::MARKER; + break; + + /* A start tag token whose tag name is one of: "marquee", "object" */ + case 'marquee': case 'object': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Insert a marker at the end of the list of active + formatting elements. */ + $this->a_formatting[] = self::MARKER; + break; + + /* A start tag token whose tag name is "xmp" */ + case 'xmp': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Switch the content model flag to the CDATA state. */ + return HTML5::CDATA; + break; + + /* A start tag whose tag name is "table" */ + case 'table': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in table". */ + $this->mode = self::IN_TABLE; + break; + + /* A start tag whose tag name is one of: "area", "basefont", + "bgsound", "br", "embed", "img", "param", "spacer", "wbr" */ + case 'area': case 'basefont': case 'bgsound': case 'br': + case 'embed': case 'img': case 'param': case 'spacer': + case 'wbr': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Immediately pop the current node off the stack of open elements. */ + array_pop($this->stack); + break; + + /* A start tag whose tag name is "hr" */ + case 'hr': + /* If the stack of open elements has a p element in scope, + then act as if an end tag with the tag name p had been seen. */ + if($this->elementInScope('p')) { + $this->emitToken(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Immediately pop the current node off the stack of open elements. */ + array_pop($this->stack); + break; + + /* A start tag whose tag name is "image" */ + case 'image': + /* Parse error. Change the token's tag name to "img" and + reprocess it. (Don't ask.) */ + $token['name'] = 'img'; + return $this->inBody($token); + break; + + /* A start tag whose tag name is "input" */ + case 'input': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an input element for the token. */ + $element = $this->insertElement($token, false); + + /* If the form element pointer is not null, then associate the + input element with the form element pointed to by the form + element pointer. */ + $this->form_pointer !== null + ? $this->form_pointer->appendChild($element) + : end($this->stack)->appendChild($element); + + /* Pop that input element off the stack of open elements. */ + array_pop($this->stack); + break; + + /* A start tag whose tag name is "isindex" */ + case 'isindex': + /* Parse error. */ + // w/e + + /* If the form element pointer is not null, + then ignore the token. */ + if($this->form_pointer === null) { + /* Act as if a start tag token with the tag name "form" had + been seen. */ + $this->inBody(array( + 'name' => 'body', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a start tag token with the tag name "hr" had + been seen. */ + $this->inBody(array( + 'name' => 'hr', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a start tag token with the tag name "p" had + been seen. */ + $this->inBody(array( + 'name' => 'p', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a start tag token with the tag name "label" + had been seen. */ + $this->inBody(array( + 'name' => 'label', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + /* Act as if a stream of character tokens had been seen. */ + $this->insertText('This is a searchable index. '. + 'Insert your search keywords here: '); + + /* Act as if a start tag token with the tag name "input" + had been seen, with all the attributes from the "isindex" + token, except with the "name" attribute set to the value + "isindex" (ignoring any explicit "name" attribute). */ + $attr = $token['attr']; + $attr[] = array('name' => 'name', 'value' => 'isindex'); + + $this->inBody(array( + 'name' => 'input', + 'type' => HTML5::STARTTAG, + 'attr' => $attr + )); + + /* Act as if a stream of character tokens had been seen + (see below for what they should say). */ + $this->insertText('This is a searchable index. '. + 'Insert your search keywords here: '); + + /* Act as if an end tag token with the tag name "label" + had been seen. */ + $this->inBody(array( + 'name' => 'label', + 'type' => HTML5::ENDTAG + )); + + /* Act as if an end tag token with the tag name "p" had + been seen. */ + $this->inBody(array( + 'name' => 'p', + 'type' => HTML5::ENDTAG + )); + + /* Act as if a start tag token with the tag name "hr" had + been seen. */ + $this->inBody(array( + 'name' => 'hr', + 'type' => HTML5::ENDTAG + )); + + /* Act as if an end tag token with the tag name "form" had + been seen. */ + $this->inBody(array( + 'name' => 'form', + 'type' => HTML5::ENDTAG + )); + } + break; + + /* A start tag whose tag name is "textarea" */ + case 'textarea': + $this->insertElement($token); + + /* Switch the tokeniser's content model flag to the + RCDATA state. */ + return HTML5::RCDATA; + break; + + /* A start tag whose tag name is one of: "iframe", "noembed", + "noframes" */ + case 'iframe': case 'noembed': case 'noframes': + $this->insertElement($token); + + /* Switch the tokeniser's content model flag to the CDATA state. */ + return HTML5::CDATA; + break; + + /* A start tag whose tag name is "select" */ + case 'select': + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Change the insertion mode to "in select". */ + $this->mode = self::IN_SELECT; + break; + + /* A start or end tag whose tag name is one of: "caption", "col", + "colgroup", "frame", "frameset", "head", "option", "optgroup", + "tbody", "td", "tfoot", "th", "thead", "tr". */ + case 'caption': case 'col': case 'colgroup': case 'frame': + case 'frameset': case 'head': case 'option': case 'optgroup': + case 'tbody': case 'td': case 'tfoot': case 'th': case 'thead': + case 'tr': + // Parse error. Ignore the token. + break; + + /* A start or end tag whose tag name is one of: "event-source", + "section", "nav", "article", "aside", "header", "footer", + "datagrid", "command" */ + case 'event-source': case 'section': case 'nav': case 'article': + case 'aside': case 'header': case 'footer': case 'datagrid': + case 'command': + // Work in progress! + break; + + /* A start tag token not covered by the previous entries */ + default: + /* Reconstruct the active formatting elements, if any. */ + $this->reconstructActiveFormattingElements(); + + $this->insertElement($token, true, true); + break; + } + break; + + case HTML5::ENDTAG: + switch($token['name']) { + /* An end tag with the tag name "body" */ + case 'body': + /* If the second element in the stack of open elements is + not a body element, this is a parse error. Ignore the token. + (innerHTML case) */ + if(count($this->stack) < 2 || $this->stack[1]->nodeName !== 'body') { + // Ignore. + + /* If the current node is not the body element, then this + is a parse error. */ + } elseif(end($this->stack)->nodeName !== 'body') { + // Parse error. + } + + /* Change the insertion mode to "after body". */ + $this->mode = self::AFTER_BODY; + break; + + /* An end tag with the tag name "html" */ + case 'html': + /* Act as if an end tag with tag name "body" had been seen, + then, if that token wasn't ignored, reprocess the current + token. */ + $this->inBody(array( + 'name' => 'body', + 'type' => HTML5::ENDTAG + )); + + return $this->afterBody($token); + break; + + /* An end tag whose tag name is one of: "address", "blockquote", + "center", "dir", "div", "dl", "fieldset", "listing", "menu", + "ol", "pre", "ul" */ + case 'address': case 'blockquote': case 'center': case 'dir': + case 'div': case 'dl': case 'fieldset': case 'listing': + case 'menu': case 'ol': case 'pre': case 'ul': + /* If the stack of open elements has an element in scope + with the same tag name as that of the token, then generate + implied end tags. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(); + + /* Now, if the current node is not an element with + the same tag name as that of the token, then this + is a parse error. */ + // w/e + + /* If the stack of open elements has an element in + scope with the same tag name as that of the token, + then pop elements from this stack until an element + with that tag name has been popped from the stack. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === $token['name']) { + $n = -1; + } + + array_pop($this->stack); + } + } + break; + + /* An end tag whose tag name is "form" */ + case 'form': + /* If the stack of open elements has an element in scope + with the same tag name as that of the token, then generate + implied end tags. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(); + + } + + if(end($this->stack)->nodeName !== $token['name']) { + /* Now, if the current node is not an element with the + same tag name as that of the token, then this is a parse + error. */ + // w/e + + } else { + /* Otherwise, if the current node is an element with + the same tag name as that of the token pop that element + from the stack. */ + array_pop($this->stack); + } + + /* In any case, set the form element pointer to null. */ + $this->form_pointer = null; + break; + + /* An end tag whose tag name is "p" */ + case 'p': + /* If the stack of open elements has a p element in scope, + then generate implied end tags, except for p elements. */ + if($this->elementInScope('p')) { + $this->generateImpliedEndTags(array('p')); + + /* If the current node is not a p element, then this is + a parse error. */ + // k + + /* If the stack of open elements has a p element in + scope, then pop elements from this stack until the stack + no longer has a p element in scope. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->elementInScope('p')) { + array_pop($this->stack); + + } else { + break; + } + } + } + break; + + /* An end tag whose tag name is "dd", "dt", or "li" */ + case 'dd': case 'dt': case 'li': + /* If the stack of open elements has an element in scope + whose tag name matches the tag name of the token, then + generate implied end tags, except for elements with the + same tag name as the token. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(array($token['name'])); + + /* If the current node is not an element with the same + tag name as the token, then this is a parse error. */ + // w/e + + /* If the stack of open elements has an element in scope + whose tag name matches the tag name of the token, then + pop elements from this stack until an element with that + tag name has been popped from the stack. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === $token['name']) { + $n = -1; + } + + array_pop($this->stack); + } + } + break; + + /* An end tag whose tag name is one of: "h1", "h2", "h3", "h4", + "h5", "h6" */ + case 'h1': case 'h2': case 'h3': case 'h4': case 'h5': case 'h6': + $elements = array('h1', 'h2', 'h3', 'h4', 'h5', 'h6'); + + /* If the stack of open elements has in scope an element whose + tag name is one of "h1", "h2", "h3", "h4", "h5", or "h6", then + generate implied end tags. */ + if($this->elementInScope($elements)) { + $this->generateImpliedEndTags(); + + /* Now, if the current node is not an element with the same + tag name as that of the token, then this is a parse error. */ + // w/e + + /* If the stack of open elements has in scope an element + whose tag name is one of "h1", "h2", "h3", "h4", "h5", or + "h6", then pop elements from the stack until an element + with one of those tag names has been popped from the stack. */ + while($this->elementInScope($elements)) { + array_pop($this->stack); + } + } + break; + + /* An end tag whose tag name is one of: "a", "b", "big", "em", + "font", "i", "nobr", "s", "small", "strike", "strong", "tt", "u" */ + case 'a': case 'b': case 'big': case 'em': case 'font': + case 'i': case 'nobr': case 's': case 'small': case 'strike': + case 'strong': case 'tt': case 'u': + /* 1. Let the formatting element be the last element in + the list of active formatting elements that: + * is between the end of the list and the last scope + marker in the list, if any, or the start of the list + otherwise, and + * has the same tag name as the token. + */ + while(true) { + for($a = count($this->a_formatting) - 1; $a >= 0; $a--) { + if($this->a_formatting[$a] === self::MARKER) { + break; + + } elseif($this->a_formatting[$a]->tagName === $token['name']) { + $formatting_element = $this->a_formatting[$a]; + $in_stack = in_array($formatting_element, $this->stack, true); + $fe_af_pos = $a; + break; + } + } + + /* If there is no such node, or, if that node is + also in the stack of open elements but the element + is not in scope, then this is a parse error. Abort + these steps. The token is ignored. */ + if(!isset($formatting_element) || ($in_stack && + !$this->elementInScope($token['name']))) { + break; + + /* Otherwise, if there is such a node, but that node + is not in the stack of open elements, then this is a + parse error; remove the element from the list, and + abort these steps. */ + } elseif(isset($formatting_element) && !$in_stack) { + unset($this->a_formatting[$fe_af_pos]); + $this->a_formatting = array_merge($this->a_formatting); + break; + } + + /* 2. Let the furthest block be the topmost node in the + stack of open elements that is lower in the stack + than the formatting element, and is not an element in + the phrasing or formatting categories. There might + not be one. */ + $fe_s_pos = array_search($formatting_element, $this->stack, true); + $length = count($this->stack); + + for($s = $fe_s_pos + 1; $s < $length; $s++) { + $category = $this->getElementCategory($this->stack[$s]->nodeName); + + if($category !== self::PHRASING && $category !== self::FORMATTING) { + $furthest_block = $this->stack[$s]; + } + } + + /* 3. If there is no furthest block, then the UA must + skip the subsequent steps and instead just pop all + the nodes from the bottom of the stack of open + elements, from the current node up to the formatting + element, and remove the formatting element from the + list of active formatting elements. */ + if(!isset($furthest_block)) { + for($n = $length - 1; $n >= $fe_s_pos; $n--) { + array_pop($this->stack); + } + + unset($this->a_formatting[$fe_af_pos]); + $this->a_formatting = array_merge($this->a_formatting); + break; + } + + /* 4. Let the common ancestor be the element + immediately above the formatting element in the stack + of open elements. */ + $common_ancestor = $this->stack[$fe_s_pos - 1]; + + /* 5. If the furthest block has a parent node, then + remove the furthest block from its parent node. */ + if($furthest_block->parentNode !== null) { + $furthest_block->parentNode->removeChild($furthest_block); + } + + /* 6. Let a bookmark note the position of the + formatting element in the list of active formatting + elements relative to the elements on either side + of it in the list. */ + $bookmark = $fe_af_pos; + + /* 7. Let node and last node be the furthest block. + Follow these steps: */ + $node = $furthest_block; + $last_node = $furthest_block; + + while(true) { + for($n = array_search($node, $this->stack, true) - 1; $n >= 0; $n--) { + /* 7.1 Let node be the element immediately + prior to node in the stack of open elements. */ + $node = $this->stack[$n]; + + /* 7.2 If node is not in the list of active + formatting elements, then remove node from + the stack of open elements and then go back + to step 1. */ + if(!in_array($node, $this->a_formatting, true)) { + unset($this->stack[$n]); + $this->stack = array_merge($this->stack); + + } else { + break; + } + } + + /* 7.3 Otherwise, if node is the formatting + element, then go to the next step in the overall + algorithm. */ + if($node === $formatting_element) { + break; + + /* 7.4 Otherwise, if last node is the furthest + block, then move the aforementioned bookmark to + be immediately after the node in the list of + active formatting elements. */ + } elseif($last_node === $furthest_block) { + $bookmark = array_search($node, $this->a_formatting, true) + 1; + } + + /* 7.5 If node has any children, perform a + shallow clone of node, replace the entry for + node in the list of active formatting elements + with an entry for the clone, replace the entry + for node in the stack of open elements with an + entry for the clone, and let node be the clone. */ + if($node->hasChildNodes()) { + $clone = $node->cloneNode(); + $s_pos = array_search($node, $this->stack, true); + $a_pos = array_search($node, $this->a_formatting, true); + + $this->stack[$s_pos] = $clone; + $this->a_formatting[$a_pos] = $clone; + $node = $clone; + } + + /* 7.6 Insert last node into node, first removing + it from its previous parent node if any. */ + if($last_node->parentNode !== null) { + $last_node->parentNode->removeChild($last_node); + } + + $node->appendChild($last_node); + + /* 7.7 Let last node be node. */ + $last_node = $node; + } + + /* 8. Insert whatever last node ended up being in + the previous step into the common ancestor node, + first removing it from its previous parent node if + any. */ + if($last_node->parentNode !== null) { + $last_node->parentNode->removeChild($last_node); + } + + $common_ancestor->appendChild($last_node); + + /* 9. Perform a shallow clone of the formatting + element. */ + $clone = $formatting_element->cloneNode(); + + /* 10. Take all of the child nodes of the furthest + block and append them to the clone created in the + last step. */ + while($furthest_block->hasChildNodes()) { + $child = $furthest_block->firstChild; + $furthest_block->removeChild($child); + $clone->appendChild($child); + } + + /* 11. Append that clone to the furthest block. */ + $furthest_block->appendChild($clone); + + /* 12. Remove the formatting element from the list + of active formatting elements, and insert the clone + into the list of active formatting elements at the + position of the aforementioned bookmark. */ + $fe_af_pos = array_search($formatting_element, $this->a_formatting, true); + unset($this->a_formatting[$fe_af_pos]); + $this->a_formatting = array_merge($this->a_formatting); + + $af_part1 = array_slice($this->a_formatting, 0, $bookmark - 1); + $af_part2 = array_slice($this->a_formatting, $bookmark, count($this->a_formatting)); + $this->a_formatting = array_merge($af_part1, array($clone), $af_part2); + + /* 13. Remove the formatting element from the stack + of open elements, and insert the clone into the stack + of open elements immediately after (i.e. in a more + deeply nested position than) the position of the + furthest block in that stack. */ + $fe_s_pos = array_search($formatting_element, $this->stack, true); + $fb_s_pos = array_search($furthest_block, $this->stack, true); + unset($this->stack[$fe_s_pos]); + + $s_part1 = array_slice($this->stack, 0, $fb_s_pos); + $s_part2 = array_slice($this->stack, $fb_s_pos + 1, count($this->stack)); + $this->stack = array_merge($s_part1, array($clone), $s_part2); + + /* 14. Jump back to step 1 in this series of steps. */ + unset($formatting_element, $fe_af_pos, $fe_s_pos, $furthest_block); + } + break; + + /* An end tag token whose tag name is one of: "button", + "marquee", "object" */ + case 'button': case 'marquee': case 'object': + /* If the stack of open elements has an element in scope whose + tag name matches the tag name of the token, then generate implied + tags. */ + if($this->elementInScope($token['name'])) { + $this->generateImpliedEndTags(); + + /* Now, if the current node is not an element with the same + tag name as the token, then this is a parse error. */ + // k + + /* Now, if the stack of open elements has an element in scope + whose tag name matches the tag name of the token, then pop + elements from the stack until that element has been popped from + the stack, and clear the list of active formatting elements up + to the last marker. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === $token['name']) { + $n = -1; + } + + array_pop($this->stack); + } + + $marker = end(array_keys($this->a_formatting, self::MARKER, true)); + + for($n = count($this->a_formatting) - 1; $n > $marker; $n--) { + array_pop($this->a_formatting); + } + } + break; + + /* Or an end tag whose tag name is one of: "area", "basefont", + "bgsound", "br", "embed", "hr", "iframe", "image", "img", + "input", "isindex", "noembed", "noframes", "param", "select", + "spacer", "table", "textarea", "wbr" */ + case 'area': case 'basefont': case 'bgsound': case 'br': + case 'embed': case 'hr': case 'iframe': case 'image': + case 'img': case 'input': case 'isindex': case 'noembed': + case 'noframes': case 'param': case 'select': case 'spacer': + case 'table': case 'textarea': case 'wbr': + // Parse error. Ignore the token. + break; + + /* An end tag token not covered by the previous entries */ + default: + for($n = count($this->stack) - 1; $n >= 0; $n--) { + /* Initialise node to be the current node (the bottommost + node of the stack). */ + $node = end($this->stack); + + /* If node has the same tag name as the end tag token, + then: */ + if($token['name'] === $node->nodeName) { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* If the tag name of the end tag token does not + match the tag name of the current node, this is a + parse error. */ + // k + + /* Pop all the nodes from the current node up to + node, including node, then stop this algorithm. */ + for($x = count($this->stack) - $n; $x >= $n; $x--) { + array_pop($this->stack); + } + + } else { + $category = $this->getElementCategory($node); + + if($category !== self::SPECIAL && $category !== self::SCOPING) { + /* Otherwise, if node is in neither the formatting + category nor the phrasing category, then this is a + parse error. Stop this algorithm. The end tag token + is ignored. */ + return false; + } + } + } + break; + } + break; + } + } + + private function inTable($token) { + $clear = array('html', 'table'); + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $text = $this->dom->createTextNode($token['data']); + end($this->stack)->appendChild($text); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + end($this->stack)->appendChild($comment); + + /* A start tag whose tag name is "caption" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'caption') { + /* Clear the stack back to a table context. */ + $this->clearStackToTableContext($clear); + + /* Insert a marker at the end of the list of active + formatting elements. */ + $this->a_formatting[] = self::MARKER; + + /* Insert an HTML element for the token, then switch the + insertion mode to "in caption". */ + $this->insertElement($token); + $this->mode = self::IN_CAPTION; + + /* A start tag whose tag name is "colgroup" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'colgroup') { + /* Clear the stack back to a table context. */ + $this->clearStackToTableContext($clear); + + /* Insert an HTML element for the token, then switch the + insertion mode to "in column group". */ + $this->insertElement($token); + $this->mode = self::IN_CGROUP; + + /* A start tag whose tag name is "col" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'col') { + $this->inTable(array( + 'name' => 'colgroup', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + $this->inColumnGroup($token); + + /* A start tag whose tag name is one of: "tbody", "tfoot", "thead" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('tbody', 'tfoot', 'thead'))) { + /* Clear the stack back to a table context. */ + $this->clearStackToTableContext($clear); + + /* Insert an HTML element for the token, then switch the insertion + mode to "in table body". */ + $this->insertElement($token); + $this->mode = self::IN_TBODY; + + /* A start tag whose tag name is one of: "td", "th", "tr" */ + } elseif($token['type'] === HTML5::STARTTAG && + in_array($token['name'], array('td', 'th', 'tr'))) { + /* Act as if a start tag token with the tag name "tbody" had been + seen, then reprocess the current token. */ + $this->inTable(array( + 'name' => 'tbody', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inTableBody($token); + + /* A start tag whose tag name is "table" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'table') { + /* Parse error. Act as if an end tag token with the tag name "table" + had been seen, then, if that token wasn't ignored, reprocess the + current token. */ + $this->inTable(array( + 'name' => 'table', + 'type' => HTML5::ENDTAG + )); + + return $this->mainPhase($token); + + /* An end tag whose tag name is "table" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'table') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + return false; + + /* Otherwise: */ + } else { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* Now, if the current node is not a table element, then this + is a parse error. */ + // w/e + + /* Pop elements from this stack until a table element has been + popped from the stack. */ + while(true) { + $current = end($this->stack)->nodeName; + array_pop($this->stack); + + if($current === 'table') { + break; + } + } + + /* Reset the insertion mode appropriately. */ + $this->resetInsertionMode(); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html", "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html', 'tbody', 'td', + 'tfoot', 'th', 'thead', 'tr'))) { + // Parse error. Ignore the token. + + /* Anything else */ + } else { + /* Parse error. Process the token as if the insertion mode was "in + body", with the following exception: */ + + /* If the current node is a table, tbody, tfoot, thead, or tr + element, then, whenever a node would be inserted into the current + node, it must instead be inserted into the foster parent element. */ + if(in_array(end($this->stack)->nodeName, + array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { + /* The foster parent element is the parent element of the last + table element in the stack of open elements, if there is a + table element and it has such a parent element. If there is no + table element in the stack of open elements (innerHTML case), + then the foster parent element is the first element in the + stack of open elements (the html element). Otherwise, if there + is a table element in the stack of open elements, but the last + table element in the stack of open elements has no parent, or + its parent node is not an element, then the foster parent + element is the element before the last table element in the + stack of open elements. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === 'table') { + $table = $this->stack[$n]; + break; + } + } + + if(isset($table) && $table->parentNode !== null) { + $this->foster_parent = $table->parentNode; + + } elseif(!isset($table)) { + $this->foster_parent = $this->stack[0]; + + } elseif(isset($table) && ($table->parentNode === null || + $table->parentNode->nodeType !== XML_ELEMENT_NODE)) { + $this->foster_parent = $this->stack[$n - 1]; + } + } + + $this->inBody($token); + } + } + + private function inCaption($token) { + /* An end tag whose tag name is "caption" */ + if($token['type'] === HTML5::ENDTAG && $token['name'] === 'caption') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore + + /* Otherwise: */ + } else { + /* Generate implied end tags. */ + $this->generateImpliedEndTags(); + + /* Now, if the current node is not a caption element, then this + is a parse error. */ + // w/e + + /* Pop elements from this stack until a caption element has + been popped from the stack. */ + while(true) { + $node = end($this->stack)->nodeName; + array_pop($this->stack); + + if($node === 'caption') { + break; + } + } + + /* Clear the list of active formatting elements up to the last + marker. */ + $this->clearTheActiveFormattingElementsUpToTheLastMarker(); + + /* Switch the insertion mode to "in table". */ + $this->mode = self::IN_TABLE; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "td", "tfoot", "th", "thead", "tr", or an end tag whose tag + name is "table" */ + } elseif(($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', + 'thead', 'tr'))) || ($token['type'] === HTML5::ENDTAG && + $token['name'] === 'table')) { + /* Parse error. Act as if an end tag with the tag name "caption" + had been seen, then, if that token wasn't ignored, reprocess the + current token. */ + $this->inCaption(array( + 'name' => 'caption', + 'type' => HTML5::ENDTAG + )); + + return $this->inTable($token); + + /* An end tag whose tag name is one of: "body", "col", "colgroup", + "html", "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'col', 'colgroup', 'html', 'tbody', 'tfoot', 'th', + 'thead', 'tr'))) { + // Parse error. Ignore the token. + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in body". */ + $this->inBody($token); + } + } + + private function inColumnGroup($token) { + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $text = $this->dom->createTextNode($token['data']); + end($this->stack)->appendChild($text); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + end($this->stack)->appendChild($comment); + + /* A start tag whose tag name is "col" */ + } elseif($token['type'] === HTML5::STARTTAG && $token['name'] === 'col') { + /* Insert a col element for the token. Immediately pop the current + node off the stack of open elements. */ + $this->insertElement($token); + array_pop($this->stack); + + /* An end tag whose tag name is "colgroup" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'colgroup') { + /* If the current node is the root html element, then this is a + parse error, ignore the token. (innerHTML case) */ + if(end($this->stack)->nodeName === 'html') { + // Ignore + + /* Otherwise, pop the current node (which will be a colgroup + element) from the stack of open elements. Switch the insertion + mode to "in table". */ + } else { + array_pop($this->stack); + $this->mode = self::IN_TABLE; + } + + /* An end tag whose tag name is "col" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'col') { + /* Parse error. Ignore the token. */ + + /* Anything else */ + } else { + /* Act as if an end tag with the tag name "colgroup" had been seen, + and then, if that token wasn't ignored, reprocess the current token. */ + $this->inColumnGroup(array( + 'name' => 'colgroup', + 'type' => HTML5::ENDTAG + )); + + return $this->inTable($token); + } + } + + private function inTableBody($token) { + $clear = array('tbody', 'tfoot', 'thead', 'html'); + + /* A start tag whose tag name is "tr" */ + if($token['type'] === HTML5::STARTTAG && $token['name'] === 'tr') { + /* Clear the stack back to a table body context. */ + $this->clearStackToTableContext($clear); + + /* Insert a tr element for the token, then switch the insertion + mode to "in row". */ + $this->insertElement($token); + $this->mode = self::IN_ROW; + + /* A start tag whose tag name is one of: "th", "td" */ + } elseif($token['type'] === HTML5::STARTTAG && + ($token['name'] === 'th' || $token['name'] === 'td')) { + /* Parse error. Act as if a start tag with the tag name "tr" had + been seen, then reprocess the current token. */ + $this->inTableBody(array( + 'name' => 'tr', + 'type' => HTML5::STARTTAG, + 'attr' => array() + )); + + return $this->inRow($token); + + /* An end tag whose tag name is one of: "tbody", "tfoot", "thead" */ + } elseif($token['type'] === HTML5::ENDTAG && + in_array($token['name'], array('tbody', 'tfoot', 'thead'))) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore + + /* Otherwise: */ + } else { + /* Clear the stack back to a table body context. */ + $this->clearStackToTableContext($clear); + + /* Pop the current node from the stack of open elements. Switch + the insertion mode to "in table". */ + array_pop($this->stack); + $this->mode = self::IN_TABLE; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "tfoot", "thead", or an end tag whose tag name is "table" */ + } elseif(($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'tfoor', 'thead'))) || + ($token['type'] === HTML5::STARTTAG && $token['name'] === 'table')) { + /* If the stack of open elements does not have a tbody, thead, or + tfoot element in table scope, this is a parse error. Ignore the + token. (innerHTML case) */ + if(!$this->elementInScope(array('tbody', 'thead', 'tfoot'), true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Clear the stack back to a table body context. */ + $this->clearStackToTableContext($clear); + + /* Act as if an end tag with the same tag name as the current + node ("tbody", "tfoot", or "thead") had been seen, then + reprocess the current token. */ + $this->inTableBody(array( + 'name' => end($this->stack)->nodeName, + 'type' => HTML5::ENDTAG + )); + + return $this->mainPhase($token); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html", "td", "th", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html', 'td', 'th', 'tr'))) { + /* Parse error. Ignore the token. */ + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in table". */ + $this->inTable($token); + } + } + + private function inRow($token) { + $clear = array('tr', 'html'); + + /* A start tag whose tag name is one of: "th", "td" */ + if($token['type'] === HTML5::STARTTAG && + ($token['name'] === 'th' || $token['name'] === 'td')) { + /* Clear the stack back to a table row context. */ + $this->clearStackToTableContext($clear); + + /* Insert an HTML element for the token, then switch the insertion + mode to "in cell". */ + $this->insertElement($token); + $this->mode = self::IN_CELL; + + /* Insert a marker at the end of the list of active formatting + elements. */ + $this->a_formatting[] = self::MARKER; + + /* An end tag whose tag name is "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'tr') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Clear the stack back to a table row context. */ + $this->clearStackToTableContext($clear); + + /* Pop the current node (which will be a tr element) from the + stack of open elements. Switch the insertion mode to "in table + body". */ + array_pop($this->stack); + $this->mode = self::IN_TBODY; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "tfoot", "thead", "tr" or an end tag whose tag name is "table" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'tfoot', 'thead', 'tr'))) { + /* Act as if an end tag with the tag name "tr" had been seen, then, + if that token wasn't ignored, reprocess the current token. */ + $this->inRow(array( + 'name' => 'tr', + 'type' => HTML5::ENDTAG + )); + + return $this->inCell($token); + + /* An end tag whose tag name is one of: "tbody", "tfoot", "thead" */ + } elseif($token['type'] === HTML5::ENDTAG && + in_array($token['name'], array('tbody', 'tfoot', 'thead'))) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Otherwise, act as if an end tag with the tag name "tr" had + been seen, then reprocess the current token. */ + $this->inRow(array( + 'name' => 'tr', + 'type' => HTML5::ENDTAG + )); + + return $this->inCell($token); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html", "td", "th" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html', 'td', 'th', 'tr'))) { + /* Parse error. Ignore the token. */ + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in table". */ + $this->inTable($token); + } + } + + private function inCell($token) { + /* An end tag whose tag name is one of: "td", "th" */ + if($token['type'] === HTML5::ENDTAG && + ($token['name'] === 'td' || $token['name'] === 'th')) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as that of the token, then this is a + parse error and the token must be ignored. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise: */ + } else { + /* Generate implied end tags, except for elements with the same + tag name as the token. */ + $this->generateImpliedEndTags(array($token['name'])); + + /* Now, if the current node is not an element with the same tag + name as the token, then this is a parse error. */ + // k + + /* Pop elements from this stack until an element with the same + tag name as the token has been popped from the stack. */ + while(true) { + $node = end($this->stack)->nodeName; + array_pop($this->stack); + + if($node === $token['name']) { + break; + } + } + + /* Clear the list of active formatting elements up to the last + marker. */ + $this->clearTheActiveFormattingElementsUpToTheLastMarker(); + + /* Switch the insertion mode to "in row". (The current node + will be a tr element at this point.) */ + $this->mode = self::IN_ROW; + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', + 'thead', 'tr'))) { + /* If the stack of open elements does not have a td or th element + in table scope, then this is a parse error; ignore the token. + (innerHTML case) */ + if(!$this->elementInScope(array('td', 'th'), true)) { + // Ignore. + + /* Otherwise, close the cell (see below) and reprocess the current + token. */ + } else { + $this->closeCell(); + return $this->inRow($token); + } + + /* A start tag whose tag name is one of: "caption", "col", "colgroup", + "tbody", "td", "tfoot", "th", "thead", "tr" */ + } elseif($token['type'] === HTML5::STARTTAG && in_array($token['name'], + array('caption', 'col', 'colgroup', 'tbody', 'td', 'tfoot', 'th', + 'thead', 'tr'))) { + /* If the stack of open elements does not have a td or th element + in table scope, then this is a parse error; ignore the token. + (innerHTML case) */ + if(!$this->elementInScope(array('td', 'th'), true)) { + // Ignore. + + /* Otherwise, close the cell (see below) and reprocess the current + token. */ + } else { + $this->closeCell(); + return $this->inRow($token); + } + + /* An end tag whose tag name is one of: "body", "caption", "col", + "colgroup", "html" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('body', 'caption', 'col', 'colgroup', 'html'))) { + /* Parse error. Ignore the token. */ + + /* An end tag whose tag name is one of: "table", "tbody", "tfoot", + "thead", "tr" */ + } elseif($token['type'] === HTML5::ENDTAG && in_array($token['name'], + array('table', 'tbody', 'tfoot', 'thead', 'tr'))) { + /* If the stack of open elements does not have an element in table + scope with the same tag name as that of the token (which can only + happen for "tbody", "tfoot" and "thead", or, in the innerHTML case), + then this is a parse error and the token must be ignored. */ + if(!$this->elementInScope($token['name'], true)) { + // Ignore. + + /* Otherwise, close the cell (see below) and reprocess the current + token. */ + } else { + $this->closeCell(); + return $this->inRow($token); + } + + /* Anything else */ + } else { + /* Process the token as if the insertion mode was "in body". */ + $this->inBody($token); + } + } + + private function inSelect($token) { + /* Handle the token as follows: */ + + /* A character token */ + if($token['type'] === HTML5::CHARACTR) { + /* Append the token's character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag token whose tag name is "option" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'option') { + /* If the current node is an option element, act as if an end tag + with the tag name "option" had been seen. */ + if(end($this->stack)->nodeName === 'option') { + $this->inSelect(array( + 'name' => 'option', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* A start tag token whose tag name is "optgroup" */ + } elseif($token['type'] === HTML5::STARTTAG && + $token['name'] === 'optgroup') { + /* If the current node is an option element, act as if an end tag + with the tag name "option" had been seen. */ + if(end($this->stack)->nodeName === 'option') { + $this->inSelect(array( + 'name' => 'option', + 'type' => HTML5::ENDTAG + )); + } + + /* If the current node is an optgroup element, act as if an end tag + with the tag name "optgroup" had been seen. */ + if(end($this->stack)->nodeName === 'optgroup') { + $this->inSelect(array( + 'name' => 'optgroup', + 'type' => HTML5::ENDTAG + )); + } + + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* An end tag token whose tag name is "optgroup" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'optgroup') { + /* First, if the current node is an option element, and the node + immediately before it in the stack of open elements is an optgroup + element, then act as if an end tag with the tag name "option" had + been seen. */ + $elements_in_stack = count($this->stack); + + if($this->stack[$elements_in_stack - 1]->nodeName === 'option' && + $this->stack[$elements_in_stack - 2]->nodeName === 'optgroup') { + $this->inSelect(array( + 'name' => 'option', + 'type' => HTML5::ENDTAG + )); + } + + /* If the current node is an optgroup element, then pop that node + from the stack of open elements. Otherwise, this is a parse error, + ignore the token. */ + if($this->stack[$elements_in_stack - 1] === 'optgroup') { + array_pop($this->stack); + } + + /* An end tag token whose tag name is "option" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'option') { + /* If the current node is an option element, then pop that node + from the stack of open elements. Otherwise, this is a parse error, + ignore the token. */ + if(end($this->stack)->nodeName === 'option') { + array_pop($this->stack); + } + + /* An end tag whose tag name is "select" */ + } elseif($token['type'] === HTML5::ENDTAG && + $token['name'] === 'select') { + /* If the stack of open elements does not have an element in table + scope with the same tag name as the token, this is a parse error. + Ignore the token. (innerHTML case) */ + if(!$this->elementInScope($token['name'], true)) { + // w/e + + /* Otherwise: */ + } else { + /* Pop elements from the stack of open elements until a select + element has been popped from the stack. */ + while(true) { + $current = end($this->stack)->nodeName; + array_pop($this->stack); + + if($current === 'select') { + break; + } + } + + /* Reset the insertion mode appropriately. */ + $this->resetInsertionMode(); + } + + /* A start tag whose tag name is "select" */ + } elseif($token['name'] === 'select' && + $token['type'] === HTML5::STARTTAG) { + /* Parse error. Act as if the token had been an end tag with the + tag name "select" instead. */ + $this->inSelect(array( + 'name' => 'select', + 'type' => HTML5::ENDTAG + )); + + /* An end tag whose tag name is one of: "caption", "table", "tbody", + "tfoot", "thead", "tr", "td", "th" */ + } elseif(in_array($token['name'], array('caption', 'table', 'tbody', + 'tfoot', 'thead', 'tr', 'td', 'th')) && $token['type'] === HTML5::ENDTAG) { + /* Parse error. */ + // w/e + + /* If the stack of open elements has an element in table scope with + the same tag name as that of the token, then act as if an end tag + with the tag name "select" had been seen, and reprocess the token. + Otherwise, ignore the token. */ + if($this->elementInScope($token['name'], true)) { + $this->inSelect(array( + 'name' => 'select', + 'type' => HTML5::ENDTAG + )); + + $this->mainPhase($token); + } + + /* Anything else */ + } else { + /* Parse error. Ignore the token. */ + } + } + + private function afterBody($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Process the token as it would be processed if the insertion mode + was "in body". */ + $this->inBody($token); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the first element in the stack of open + elements (the html element), with the data attribute set to the + data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + $this->stack[0]->appendChild($comment); + + /* An end tag with the tag name "html" */ + } elseif($token['type'] === HTML5::ENDTAG && $token['name'] === 'html') { + /* If the parser was originally created in order to handle the + setting of an element's innerHTML attribute, this is a parse error; + ignore the token. (The element will be an html element in this + case.) (innerHTML case) */ + + /* Otherwise, switch to the trailing end phase. */ + $this->phase = self::END_PHASE; + + /* Anything else */ + } else { + /* Parse error. Set the insertion mode to "in body" and reprocess + the token. */ + $this->mode = self::IN_BODY; + return $this->inBody($token); + } + } + + private function inFrameset($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + U+000D CARRIAGE RETURN (CR), or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* A start tag with the tag name "frameset" */ + } elseif($token['name'] === 'frameset' && + $token['type'] === HTML5::STARTTAG) { + $this->insertElement($token); + + /* An end tag with the tag name "frameset" */ + } elseif($token['name'] === 'frameset' && + $token['type'] === HTML5::ENDTAG) { + /* If the current node is the root html element, then this is a + parse error; ignore the token. (innerHTML case) */ + if(end($this->stack)->nodeName === 'html') { + // Ignore + + } else { + /* Otherwise, pop the current node from the stack of open + elements. */ + array_pop($this->stack); + + /* If the parser was not originally created in order to handle + the setting of an element's innerHTML attribute (innerHTML case), + and the current node is no longer a frameset element, then change + the insertion mode to "after frameset". */ + $this->mode = self::AFTR_FRAME; + } + + /* A start tag with the tag name "frame" */ + } elseif($token['name'] === 'frame' && + $token['type'] === HTML5::STARTTAG) { + /* Insert an HTML element for the token. */ + $this->insertElement($token); + + /* Immediately pop the current node off the stack of open elements. */ + array_pop($this->stack); + + /* A start tag with the tag name "noframes" */ + } elseif($token['name'] === 'noframes' && + $token['type'] === HTML5::STARTTAG) { + /* Process the token as if the insertion mode had been "in body". */ + $this->inBody($token); + + /* Anything else */ + } else { + /* Parse error. Ignore the token. */ + } + } + + private function afterFrameset($token) { + /* Handle the token as follows: */ + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + U+000D CARRIAGE RETURN (CR), or U+0020 SPACE */ + if($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Append the character to the current node. */ + $this->insertText($token['data']); + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the current node with the data + attribute set to the data given in the comment token. */ + $this->insertComment($token['data']); + + /* An end tag with the tag name "html" */ + } elseif($token['name'] === 'html' && + $token['type'] === HTML5::ENDTAG) { + /* Switch to the trailing end phase. */ + $this->phase = self::END_PHASE; + + /* A start tag with the tag name "noframes" */ + } elseif($token['name'] === 'noframes' && + $token['type'] === HTML5::STARTTAG) { + /* Process the token as if the insertion mode had been "in body". */ + $this->inBody($token); + + /* Anything else */ + } else { + /* Parse error. Ignore the token. */ + } + } + + private function trailingEndPhase($token) { + /* After the main phase, as each token is emitted from the tokenisation + stage, it must be processed as described in this section. */ + + /* A DOCTYPE token */ + if($token['type'] === HTML5::DOCTYPE) { + // Parse error. Ignore the token. + + /* A comment token */ + } elseif($token['type'] === HTML5::COMMENT) { + /* Append a Comment node to the Document object with the data + attribute set to the data given in the comment token. */ + $comment = $this->dom->createComment($token['data']); + $this->dom->appendChild($comment); + + /* A character token that is one of one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE */ + } elseif($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) { + /* Process the token as it would be processed in the main phase. */ + $this->mainPhase($token); + + /* A character token that is not one of U+0009 CHARACTER TABULATION, + U+000A LINE FEED (LF), U+000B LINE TABULATION, U+000C FORM FEED (FF), + or U+0020 SPACE. Or a start tag token. Or an end tag token. */ + } elseif(($token['type'] === HTML5::CHARACTR && + preg_match('/^[\t\n\x0b\x0c ]+$/', $token['data'])) || + $token['type'] === HTML5::STARTTAG || $token['type'] === HTML5::ENDTAG) { + /* Parse error. Switch back to the main phase and reprocess the + token. */ + $this->phase = self::MAIN_PHASE; + return $this->mainPhase($token); + + /* An end-of-file token */ + } elseif($token['type'] === HTML5::EOF) { + /* OMG DONE!! */ + } + } + + private function insertElement($token, $append = true, $check = false) { + // Proprietary workaround for libxml2's limitations with tag names + if ($check) { + // Slightly modified HTML5 tag-name modification, + // removing anything that's not an ASCII letter, digit, or hyphen + $token['name'] = preg_replace('/[^a-z0-9-]/i', '', $token['name']); + // Remove leading hyphens and numbers + $token['name'] = ltrim($token['name'], '-0..9'); + // In theory, this should ever be needed, but just in case + if ($token['name'] === '') $token['name'] = 'span'; // arbitrary generic choice + } + + $el = $this->dom->createElement($token['name']); + + foreach($token['attr'] as $attr) { + if(!$el->hasAttribute($attr['name'])) { + $el->setAttribute($attr['name'], $attr['value']); + } + } + + $this->appendToRealParent($el); + $this->stack[] = $el; + + return $el; + } + + private function insertText($data) { + $text = $this->dom->createTextNode($data); + $this->appendToRealParent($text); + } + + private function insertComment($data) { + $comment = $this->dom->createComment($data); + $this->appendToRealParent($comment); + } + + private function appendToRealParent($node) { + if($this->foster_parent === null) { + end($this->stack)->appendChild($node); + + } elseif($this->foster_parent !== null) { + /* If the foster parent element is the parent element of the + last table element in the stack of open elements, then the new + node must be inserted immediately before the last table element + in the stack of open elements in the foster parent element; + otherwise, the new node must be appended to the foster parent + element. */ + for($n = count($this->stack) - 1; $n >= 0; $n--) { + if($this->stack[$n]->nodeName === 'table' && + $this->stack[$n]->parentNode !== null) { + $table = $this->stack[$n]; + break; + } + } + + if(isset($table) && $this->foster_parent->isSameNode($table->parentNode)) + $this->foster_parent->insertBefore($node, $table); + else + $this->foster_parent->appendChild($node); + + $this->foster_parent = null; + } + } + + private function elementInScope($el, $table = false) { + if(is_array($el)) { + foreach($el as $element) { + if($this->elementInScope($element, $table)) { + return true; + } + } + + return false; + } + + $leng = count($this->stack); + + for($n = 0; $n < $leng; $n++) { + /* 1. Initialise node to be the current node (the bottommost node of + the stack). */ + $node = $this->stack[$leng - 1 - $n]; + + if($node->tagName === $el) { + /* 2. If node is the target node, terminate in a match state. */ + return true; + + } elseif($node->tagName === 'table') { + /* 3. Otherwise, if node is a table element, terminate in a failure + state. */ + return false; + + } elseif($table === true && in_array($node->tagName, array('caption', 'td', + 'th', 'button', 'marquee', 'object'))) { + /* 4. Otherwise, if the algorithm is the "has an element in scope" + variant (rather than the "has an element in table scope" variant), + and node is one of the following, terminate in a failure state. */ + return false; + + } elseif($node === $node->ownerDocument->documentElement) { + /* 5. Otherwise, if node is an html element (root element), terminate + in a failure state. (This can only happen if the node is the topmost + node of the stack of open elements, and prevents the next step from + being invoked if there are no more elements in the stack.) */ + return false; + } + + /* Otherwise, set node to the previous entry in the stack of open + elements and return to step 2. (This will never fail, since the loop + will always terminate in the previous step if the top of the stack + is reached.) */ + } + } + + private function reconstructActiveFormattingElements() { + /* 1. If there are no entries in the list of active formatting elements, + then there is nothing to reconstruct; stop this algorithm. */ + $formatting_elements = count($this->a_formatting); + + if($formatting_elements === 0) { + return false; + } + + /* 3. Let entry be the last (most recently added) element in the list + of active formatting elements. */ + $entry = end($this->a_formatting); + + /* 2. If the last (most recently added) entry in the list of active + formatting elements is a marker, or if it is an element that is in the + stack of open elements, then there is nothing to reconstruct; stop this + algorithm. */ + if($entry === self::MARKER || in_array($entry, $this->stack, true)) { + return false; + } + + for($a = $formatting_elements - 1; $a >= 0; true) { + /* 4. If there are no entries before entry in the list of active + formatting elements, then jump to step 8. */ + if($a === 0) { + $step_seven = false; + break; + } + + /* 5. Let entry be the entry one earlier than entry in the list of + active formatting elements. */ + $a--; + $entry = $this->a_formatting[$a]; + + /* 6. If entry is neither a marker nor an element that is also in + thetack of open elements, go to step 4. */ + if($entry === self::MARKER || in_array($entry, $this->stack, true)) { + break; + } + } + + while(true) { + /* 7. Let entry be the element one later than entry in the list of + active formatting elements. */ + if(isset($step_seven) && $step_seven === true) { + $a++; + $entry = $this->a_formatting[$a]; + } + + /* 8. Perform a shallow clone of the element entry to obtain clone. */ + $clone = $entry->cloneNode(); + + /* 9. Append clone to the current node and push it onto the stack + of open elements so that it is the new current node. */ + end($this->stack)->appendChild($clone); + $this->stack[] = $clone; + + /* 10. Replace the entry for entry in the list with an entry for + clone. */ + $this->a_formatting[$a] = $clone; + + /* 11. If the entry for clone in the list of active formatting + elements is not the last entry in the list, return to step 7. */ + if(end($this->a_formatting) !== $clone) { + $step_seven = true; + } else { + break; + } + } + } + + private function clearTheActiveFormattingElementsUpToTheLastMarker() { + /* When the steps below require the UA to clear the list of active + formatting elements up to the last marker, the UA must perform the + following steps: */ + + while(true) { + /* 1. Let entry be the last (most recently added) entry in the list + of active formatting elements. */ + $entry = end($this->a_formatting); + + /* 2. Remove entry from the list of active formatting elements. */ + array_pop($this->a_formatting); + + /* 3. If entry was a marker, then stop the algorithm at this point. + The list has been cleared up to the last marker. */ + if($entry === self::MARKER) { + break; + } + } + } + + private function generateImpliedEndTags($exclude = array()) { + /* When the steps below require the UA to generate implied end tags, + then, if the current node is a dd element, a dt element, an li element, + a p element, a td element, a th element, or a tr element, the UA must + act as if an end tag with the respective tag name had been seen and + then generate implied end tags again. */ + $node = end($this->stack); + $elements = array_diff(array('dd', 'dt', 'li', 'p', 'td', 'th', 'tr'), $exclude); + + while(in_array(end($this->stack)->nodeName, $elements)) { + array_pop($this->stack); + } + } + + private function getElementCategory($node) { + $name = $node->tagName; + if(in_array($name, $this->special)) + return self::SPECIAL; + + elseif(in_array($name, $this->scoping)) + return self::SCOPING; + + elseif(in_array($name, $this->formatting)) + return self::FORMATTING; + + else + return self::PHRASING; + } + + private function clearStackToTableContext($elements) { + /* When the steps above require the UA to clear the stack back to a + table context, it means that the UA must, while the current node is not + a table element or an html element, pop elements from the stack of open + elements. If this causes any elements to be popped from the stack, then + this is a parse error. */ + while(true) { + $node = end($this->stack)->nodeName; + + if(in_array($node, $elements)) { + break; + } else { + array_pop($this->stack); + } + } + } + + private function resetInsertionMode() { + /* 1. Let last be false. */ + $last = false; + $leng = count($this->stack); + + for($n = $leng - 1; $n >= 0; $n--) { + /* 2. Let node be the last node in the stack of open elements. */ + $node = $this->stack[$n]; + + /* 3. If node is the first node in the stack of open elements, then + set last to true. If the element whose innerHTML attribute is being + set is neither a td element nor a th element, then set node to the + element whose innerHTML attribute is being set. (innerHTML case) */ + if($this->stack[0]->isSameNode($node)) { + $last = true; + } + + /* 4. If node is a select element, then switch the insertion mode to + "in select" and abort these steps. (innerHTML case) */ + if($node->nodeName === 'select') { + $this->mode = self::IN_SELECT; + break; + + /* 5. If node is a td or th element, then switch the insertion mode + to "in cell" and abort these steps. */ + } elseif($node->nodeName === 'td' || $node->nodeName === 'th') { + $this->mode = self::IN_CELL; + break; + + /* 6. If node is a tr element, then switch the insertion mode to + "in row" and abort these steps. */ + } elseif($node->nodeName === 'tr') { + $this->mode = self::IN_ROW; + break; + + /* 7. If node is a tbody, thead, or tfoot element, then switch the + insertion mode to "in table body" and abort these steps. */ + } elseif(in_array($node->nodeName, array('tbody', 'thead', 'tfoot'))) { + $this->mode = self::IN_TBODY; + break; + + /* 8. If node is a caption element, then switch the insertion mode + to "in caption" and abort these steps. */ + } elseif($node->nodeName === 'caption') { + $this->mode = self::IN_CAPTION; + break; + + /* 9. If node is a colgroup element, then switch the insertion mode + to "in column group" and abort these steps. (innerHTML case) */ + } elseif($node->nodeName === 'colgroup') { + $this->mode = self::IN_CGROUP; + break; + + /* 10. If node is a table element, then switch the insertion mode + to "in table" and abort these steps. */ + } elseif($node->nodeName === 'table') { + $this->mode = self::IN_TABLE; + break; + + /* 11. If node is a head element, then switch the insertion mode + to "in body" ("in body"! not "in head"!) and abort these steps. + (innerHTML case) */ + } elseif($node->nodeName === 'head') { + $this->mode = self::IN_BODY; + break; + + /* 12. If node is a body element, then switch the insertion mode to + "in body" and abort these steps. */ + } elseif($node->nodeName === 'body') { + $this->mode = self::IN_BODY; + break; + + /* 13. If node is a frameset element, then switch the insertion + mode to "in frameset" and abort these steps. (innerHTML case) */ + } elseif($node->nodeName === 'frameset') { + $this->mode = self::IN_FRAME; + break; + + /* 14. If node is an html element, then: if the head element + pointer is null, switch the insertion mode to "before head", + otherwise, switch the insertion mode to "after head". In either + case, abort these steps. (innerHTML case) */ + } elseif($node->nodeName === 'html') { + $this->mode = ($this->head_pointer === null) + ? self::BEFOR_HEAD + : self::AFTER_HEAD; + + break; + + /* 15. If last is true, then set the insertion mode to "in body" + and abort these steps. (innerHTML case) */ + } elseif($last) { + $this->mode = self::IN_BODY; + break; + } + } + } + + private function closeCell() { + /* If the stack of open elements has a td or th element in table scope, + then act as if an end tag token with that tag name had been seen. */ + foreach(array('td', 'th') as $cell) { + if($this->elementInScope($cell, true)) { + $this->inCell(array( + 'name' => $cell, + 'type' => HTML5::ENDTAG + )); + + break; + } + } + } + + public function save() { + return $this->dom; + } +} +?> diff --git a/lib/htmlpurifier/library/HTMLPurifier/PercentEncoder.php b/lib/htmlpurifier/library/HTMLPurifier/PercentEncoder.php new file mode 100644 index 000000000..a43c44f4c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/PercentEncoder.php @@ -0,0 +1,98 @@ +<?php + +/** + * Class that handles operations involving percent-encoding in URIs. + * + * @warning + * Be careful when reusing instances of PercentEncoder. The object + * you use for normalize() SHOULD NOT be used for encode(), or + * vice-versa. + */ +class HTMLPurifier_PercentEncoder +{ + + /** + * Reserved characters to preserve when using encode(). + */ + protected $preserve = array(); + + /** + * String of characters that should be preserved while using encode(). + */ + public function __construct($preserve = false) { + // unreserved letters, ought to const-ify + for ($i = 48; $i <= 57; $i++) $this->preserve[$i] = true; // digits + for ($i = 65; $i <= 90; $i++) $this->preserve[$i] = true; // upper-case + for ($i = 97; $i <= 122; $i++) $this->preserve[$i] = true; // lower-case + $this->preserve[45] = true; // Dash - + $this->preserve[46] = true; // Period . + $this->preserve[95] = true; // Underscore _ + $this->preserve[126]= true; // Tilde ~ + + // extra letters not to escape + if ($preserve !== false) { + for ($i = 0, $c = strlen($preserve); $i < $c; $i++) { + $this->preserve[ord($preserve[$i])] = true; + } + } + } + + /** + * Our replacement for urlencode, it encodes all non-reserved characters, + * as well as any extra characters that were instructed to be preserved. + * @note + * Assumes that the string has already been normalized, making any + * and all percent escape sequences valid. Percents will not be + * re-escaped, regardless of their status in $preserve + * @param $string String to be encoded + * @return Encoded string. + */ + public function encode($string) { + $ret = ''; + for ($i = 0, $c = strlen($string); $i < $c; $i++) { + if ($string[$i] !== '%' && !isset($this->preserve[$int = ord($string[$i])]) ) { + $ret .= '%' . sprintf('%02X', $int); + } else { + $ret .= $string[$i]; + } + } + return $ret; + } + + /** + * Fix up percent-encoding by decoding unreserved characters and normalizing. + * @warning This function is affected by $preserve, even though the + * usual desired behavior is for this not to preserve those + * characters. Be careful when reusing instances of PercentEncoder! + * @param $string String to normalize + */ + public function normalize($string) { + if ($string == '') return ''; + $parts = explode('%', $string); + $ret = array_shift($parts); + foreach ($parts as $part) { + $length = strlen($part); + if ($length < 2) { + $ret .= '%25' . $part; + continue; + } + $encoding = substr($part, 0, 2); + $text = substr($part, 2); + if (!ctype_xdigit($encoding)) { + $ret .= '%25' . $part; + continue; + } + $int = hexdec($encoding); + if (isset($this->preserve[$int])) { + $ret .= chr($int) . $text; + continue; + } + $encoding = strtoupper($encoding); + $ret .= '%' . $encoding . $text; + } + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer.php b/lib/htmlpurifier/library/HTMLPurifier/Printer.php new file mode 100644 index 000000000..e7eb82e83 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer.php @@ -0,0 +1,176 @@ +<?php + +// OUT OF DATE, NEEDS UPDATING! +// USE XMLWRITER! + +class HTMLPurifier_Printer +{ + + /** + * Instance of HTMLPurifier_Generator for HTML generation convenience funcs + */ + protected $generator; + + /** + * Instance of HTMLPurifier_Config, for easy access + */ + protected $config; + + /** + * Initialize $generator. + */ + public function __construct() { + } + + /** + * Give generator necessary configuration if possible + */ + public function prepareGenerator($config) { + $all = $config->getAll(); + $context = new HTMLPurifier_Context(); + $this->generator = new HTMLPurifier_Generator($config, $context); + } + + /** + * Main function that renders object or aspect of that object + * @note Parameters vary depending on printer + */ + // function render() {} + + /** + * Returns a start tag + * @param $tag Tag name + * @param $attr Attribute array + */ + protected function start($tag, $attr = array()) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_Start($tag, $attr ? $attr : array()) + ); + } + + /** + * Returns an end teg + * @param $tag Tag name + */ + protected function end($tag) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_End($tag) + ); + } + + /** + * Prints a complete element with content inside + * @param $tag Tag name + * @param $contents Element contents + * @param $attr Tag attributes + * @param $escape Bool whether or not to escape contents + */ + protected function element($tag, $contents, $attr = array(), $escape = true) { + return $this->start($tag, $attr) . + ($escape ? $this->escape($contents) : $contents) . + $this->end($tag); + } + + protected function elementEmpty($tag, $attr = array()) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_Empty($tag, $attr) + ); + } + + protected function text($text) { + return $this->generator->generateFromToken( + new HTMLPurifier_Token_Text($text) + ); + } + + /** + * Prints a simple key/value row in a table. + * @param $name Key + * @param $value Value + */ + protected function row($name, $value) { + if (is_bool($value)) $value = $value ? 'On' : 'Off'; + return + $this->start('tr') . "\n" . + $this->element('th', $name) . "\n" . + $this->element('td', $value) . "\n" . + $this->end('tr') + ; + } + + /** + * Escapes a string for HTML output. + * @param $string String to escape + */ + protected function escape($string) { + $string = HTMLPurifier_Encoder::cleanUTF8($string); + $string = htmlspecialchars($string, ENT_COMPAT, 'UTF-8'); + return $string; + } + + /** + * Takes a list of strings and turns them into a single list + * @param $array List of strings + * @param $polite Bool whether or not to add an end before the last + */ + protected function listify($array, $polite = false) { + if (empty($array)) return 'None'; + $ret = ''; + $i = count($array); + foreach ($array as $value) { + $i--; + $ret .= $value; + if ($i > 0 && !($polite && $i == 1)) $ret .= ', '; + if ($polite && $i == 1) $ret .= 'and '; + } + return $ret; + } + + /** + * Retrieves the class of an object without prefixes, as well as metadata + * @param $obj Object to determine class of + * @param $prefix Further prefix to remove + */ + protected function getClass($obj, $sec_prefix = '') { + static $five = null; + if ($five === null) $five = version_compare(PHP_VERSION, '5', '>='); + $prefix = 'HTMLPurifier_' . $sec_prefix; + if (!$five) $prefix = strtolower($prefix); + $class = str_replace($prefix, '', get_class($obj)); + $lclass = strtolower($class); + $class .= '('; + switch ($lclass) { + case 'enum': + $values = array(); + foreach ($obj->valid_values as $value => $bool) { + $values[] = $value; + } + $class .= implode(', ', $values); + break; + case 'css_composite': + $values = array(); + foreach ($obj->defs as $def) { + $values[] = $this->getClass($def, $sec_prefix); + } + $class .= implode(', ', $values); + break; + case 'css_multiple': + $class .= $this->getClass($obj->single, $sec_prefix) . ', '; + $class .= $obj->max; + break; + case 'css_denyelementdecorator': + $class .= $this->getClass($obj->def, $sec_prefix) . ', '; + $class .= $obj->element; + break; + case 'css_importantdecorator': + $class .= $this->getClass($obj->def, $sec_prefix); + if ($obj->allow) $class .= ', !important'; + break; + } + $class .= ')'; + return $class; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/CSSDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/Printer/CSSDefinition.php new file mode 100644 index 000000000..81f986590 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/CSSDefinition.php @@ -0,0 +1,38 @@ +<?php + +class HTMLPurifier_Printer_CSSDefinition extends HTMLPurifier_Printer +{ + + protected $def; + + public function render($config) { + $this->def = $config->getCSSDefinition(); + $ret = ''; + + $ret .= $this->start('div', array('class' => 'HTMLPurifier_Printer')); + $ret .= $this->start('table'); + + $ret .= $this->element('caption', 'Properties ($info)'); + + $ret .= $this->start('thead'); + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Property', array('class' => 'heavy')); + $ret .= $this->element('th', 'Definition', array('class' => 'heavy', 'style' => 'width:auto;')); + $ret .= $this->end('tr'); + $ret .= $this->end('thead'); + + ksort($this->def->info); + foreach ($this->def->info as $property => $obj) { + $name = $this->getClass($obj, 'AttrDef_'); + $ret .= $this->row($property, $name); + } + + $ret .= $this->end('table'); + $ret .= $this->end('div'); + + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.css b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.css new file mode 100644 index 000000000..3ff1a88aa --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.css @@ -0,0 +1,10 @@ + +.hp-config {} + +.hp-config tbody th {text-align:right; padding-right:0.5em;} +.hp-config thead, .hp-config .namespace {background:#3C578C; color:#FFF;} +.hp-config .namespace th {text-align:center;} +.hp-config .verbose {display:none;} +.hp-config .controls {text-align:center;} + +/* vim: et sw=4 sts=4 */ diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.js b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.js new file mode 100644 index 000000000..cba00c9b8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.js @@ -0,0 +1,5 @@ +function toggleWriteability(id_of_patient, checked) { + document.getElementById(id_of_patient).disabled = checked; +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.php b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.php new file mode 100644 index 000000000..02aa65689 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/ConfigForm.php @@ -0,0 +1,368 @@ +<?php + +/** + * @todo Rewrite to use Interchange objects + */ +class HTMLPurifier_Printer_ConfigForm extends HTMLPurifier_Printer +{ + + /** + * Printers for specific fields + */ + protected $fields = array(); + + /** + * Documentation URL, can have fragment tagged on end + */ + protected $docURL; + + /** + * Name of form element to stuff config in + */ + protected $name; + + /** + * Whether or not to compress directive names, clipping them off + * after a certain amount of letters. False to disable or integer letters + * before clipping. + */ + protected $compress = false; + + /** + * @param $name Form element name for directives to be stuffed into + * @param $doc_url String documentation URL, will have fragment tagged on + * @param $compress Integer max length before compressing a directive name, set to false to turn off + */ + public function __construct( + $name, $doc_url = null, $compress = false + ) { + parent::__construct(); + $this->docURL = $doc_url; + $this->name = $name; + $this->compress = $compress; + // initialize sub-printers + $this->fields[0] = new HTMLPurifier_Printer_ConfigForm_default(); + $this->fields[HTMLPurifier_VarParser::BOOL] = new HTMLPurifier_Printer_ConfigForm_bool(); + } + + /** + * Sets default column and row size for textareas in sub-printers + * @param $cols Integer columns of textarea, null to use default + * @param $rows Integer rows of textarea, null to use default + */ + public function setTextareaDimensions($cols = null, $rows = null) { + if ($cols) $this->fields['default']->cols = $cols; + if ($rows) $this->fields['default']->rows = $rows; + } + + /** + * Retrieves styling, in case it is not accessible by webserver + */ + public static function getCSS() { + return file_get_contents(HTMLPURIFIER_PREFIX . '/HTMLPurifier/Printer/ConfigForm.css'); + } + + /** + * Retrieves JavaScript, in case it is not accessible by webserver + */ + public static function getJavaScript() { + return file_get_contents(HTMLPURIFIER_PREFIX . '/HTMLPurifier/Printer/ConfigForm.js'); + } + + /** + * Returns HTML output for a configuration form + * @param $config Configuration object of current form state, or an array + * where [0] has an HTML namespace and [1] is being rendered. + * @param $allowed Optional namespace(s) and directives to restrict form to. + */ + public function render($config, $allowed = true, $render_controls = true) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + + $this->config = $config; + $this->genConfig = $gen_config; + $this->prepareGenerator($gen_config); + + $allowed = HTMLPurifier_Config::getAllowedDirectivesForForm($allowed, $config->def); + $all = array(); + foreach ($allowed as $key) { + list($ns, $directive) = $key; + $all[$ns][$directive] = $config->get($ns .'.'. $directive); + } + + $ret = ''; + $ret .= $this->start('table', array('class' => 'hp-config')); + $ret .= $this->start('thead'); + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Directive', array('class' => 'hp-directive')); + $ret .= $this->element('th', 'Value', array('class' => 'hp-value')); + $ret .= $this->end('tr'); + $ret .= $this->end('thead'); + foreach ($all as $ns => $directives) { + $ret .= $this->renderNamespace($ns, $directives); + } + if ($render_controls) { + $ret .= $this->start('tbody'); + $ret .= $this->start('tr'); + $ret .= $this->start('td', array('colspan' => 2, 'class' => 'controls')); + $ret .= $this->elementEmpty('input', array('type' => 'submit', 'value' => 'Submit')); + $ret .= '[<a href="?">Reset</a>]'; + $ret .= $this->end('td'); + $ret .= $this->end('tr'); + $ret .= $this->end('tbody'); + } + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders a single namespace + * @param $ns String namespace name + * @param $directive Associative array of directives to values + */ + protected function renderNamespace($ns, $directives) { + $ret = ''; + $ret .= $this->start('tbody', array('class' => 'namespace')); + $ret .= $this->start('tr'); + $ret .= $this->element('th', $ns, array('colspan' => 2)); + $ret .= $this->end('tr'); + $ret .= $this->end('tbody'); + $ret .= $this->start('tbody'); + foreach ($directives as $directive => $value) { + $ret .= $this->start('tr'); + $ret .= $this->start('th'); + if ($this->docURL) { + $url = str_replace('%s', urlencode("$ns.$directive"), $this->docURL); + $ret .= $this->start('a', array('href' => $url)); + } + $attr = array('for' => "{$this->name}:$ns.$directive"); + + // crop directive name if it's too long + if (!$this->compress || (strlen($directive) < $this->compress)) { + $directive_disp = $directive; + } else { + $directive_disp = substr($directive, 0, $this->compress - 2) . '...'; + $attr['title'] = $directive; + } + + $ret .= $this->element( + 'label', + $directive_disp, + // component printers must create an element with this id + $attr + ); + if ($this->docURL) $ret .= $this->end('a'); + $ret .= $this->end('th'); + + $ret .= $this->start('td'); + $def = $this->config->def->info["$ns.$directive"]; + if (is_int($def)) { + $allow_null = $def < 0; + $type = abs($def); + } else { + $type = $def->type; + $allow_null = isset($def->allow_null); + } + if (!isset($this->fields[$type])) $type = 0; // default + $type_obj = $this->fields[$type]; + if ($allow_null) { + $type_obj = new HTMLPurifier_Printer_ConfigForm_NullDecorator($type_obj); + } + $ret .= $type_obj->render($ns, $directive, $value, $this->name, array($this->genConfig, $this->config)); + $ret .= $this->end('td'); + $ret .= $this->end('tr'); + } + $ret .= $this->end('tbody'); + return $ret; + } + +} + +/** + * Printer decorator for directives that accept null + */ +class HTMLPurifier_Printer_ConfigForm_NullDecorator extends HTMLPurifier_Printer { + /** + * Printer being decorated + */ + protected $obj; + /** + * @param $obj Printer to decorate + */ + public function __construct($obj) { + parent::__construct(); + $this->obj = $obj; + } + public function render($ns, $directive, $value, $name, $config) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + $this->prepareGenerator($gen_config); + + $ret = ''; + $ret .= $this->start('label', array('for' => "$name:Null_$ns.$directive")); + $ret .= $this->element('span', "$ns.$directive:", array('class' => 'verbose')); + $ret .= $this->text(' Null/Disabled'); + $ret .= $this->end('label'); + $attr = array( + 'type' => 'checkbox', + 'value' => '1', + 'class' => 'null-toggle', + 'name' => "$name"."[Null_$ns.$directive]", + 'id' => "$name:Null_$ns.$directive", + 'onclick' => "toggleWriteability('$name:$ns.$directive',checked)" // INLINE JAVASCRIPT!!!! + ); + if ($this->obj instanceof HTMLPurifier_Printer_ConfigForm_bool) { + // modify inline javascript slightly + $attr['onclick'] = "toggleWriteability('$name:Yes_$ns.$directive',checked);toggleWriteability('$name:No_$ns.$directive',checked)"; + } + if ($value === null) $attr['checked'] = 'checked'; + $ret .= $this->elementEmpty('input', $attr); + $ret .= $this->text(' or '); + $ret .= $this->elementEmpty('br'); + $ret .= $this->obj->render($ns, $directive, $value, $name, array($gen_config, $config)); + return $ret; + } +} + +/** + * Swiss-army knife configuration form field printer + */ +class HTMLPurifier_Printer_ConfigForm_default extends HTMLPurifier_Printer { + public $cols = 18; + public $rows = 5; + public function render($ns, $directive, $value, $name, $config) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + $this->prepareGenerator($gen_config); + // this should probably be split up a little + $ret = ''; + $def = $config->def->info["$ns.$directive"]; + if (is_int($def)) { + $type = abs($def); + } else { + $type = $def->type; + } + if (is_array($value)) { + switch ($type) { + case HTMLPurifier_VarParser::LOOKUP: + $array = $value; + $value = array(); + foreach ($array as $val => $b) { + $value[] = $val; + } + case HTMLPurifier_VarParser::ALIST: + $value = implode(PHP_EOL, $value); + break; + case HTMLPurifier_VarParser::HASH: + $nvalue = ''; + foreach ($value as $i => $v) { + $nvalue .= "$i:$v" . PHP_EOL; + } + $value = $nvalue; + break; + default: + $value = ''; + } + } + if ($type === HTMLPurifier_VarParser::MIXED) { + return 'Not supported'; + $value = serialize($value); + } + $attr = array( + 'name' => "$name"."[$ns.$directive]", + 'id' => "$name:$ns.$directive" + ); + if ($value === null) $attr['disabled'] = 'disabled'; + if (isset($def->allowed)) { + $ret .= $this->start('select', $attr); + foreach ($def->allowed as $val => $b) { + $attr = array(); + if ($value == $val) $attr['selected'] = 'selected'; + $ret .= $this->element('option', $val, $attr); + } + $ret .= $this->end('select'); + } elseif ( + $type === HTMLPurifier_VarParser::TEXT || + $type === HTMLPurifier_VarParser::ITEXT || + $type === HTMLPurifier_VarParser::ALIST || + $type === HTMLPurifier_VarParser::HASH || + $type === HTMLPurifier_VarParser::LOOKUP + ) { + $attr['cols'] = $this->cols; + $attr['rows'] = $this->rows; + $ret .= $this->start('textarea', $attr); + $ret .= $this->text($value); + $ret .= $this->end('textarea'); + } else { + $attr['value'] = $value; + $attr['type'] = 'text'; + $ret .= $this->elementEmpty('input', $attr); + } + return $ret; + } +} + +/** + * Bool form field printer + */ +class HTMLPurifier_Printer_ConfigForm_bool extends HTMLPurifier_Printer { + public function render($ns, $directive, $value, $name, $config) { + if (is_array($config) && isset($config[0])) { + $gen_config = $config[0]; + $config = $config[1]; + } else { + $gen_config = $config; + } + $this->prepareGenerator($gen_config); + $ret = ''; + $ret .= $this->start('div', array('id' => "$name:$ns.$directive")); + + $ret .= $this->start('label', array('for' => "$name:Yes_$ns.$directive")); + $ret .= $this->element('span', "$ns.$directive:", array('class' => 'verbose')); + $ret .= $this->text(' Yes'); + $ret .= $this->end('label'); + + $attr = array( + 'type' => 'radio', + 'name' => "$name"."[$ns.$directive]", + 'id' => "$name:Yes_$ns.$directive", + 'value' => '1' + ); + if ($value === true) $attr['checked'] = 'checked'; + if ($value === null) $attr['disabled'] = 'disabled'; + $ret .= $this->elementEmpty('input', $attr); + + $ret .= $this->start('label', array('for' => "$name:No_$ns.$directive")); + $ret .= $this->element('span', "$ns.$directive:", array('class' => 'verbose')); + $ret .= $this->text(' No'); + $ret .= $this->end('label'); + + $attr = array( + 'type' => 'radio', + 'name' => "$name"."[$ns.$directive]", + 'id' => "$name:No_$ns.$directive", + 'value' => '0' + ); + if ($value === false) $attr['checked'] = 'checked'; + if ($value === null) $attr['disabled'] = 'disabled'; + $ret .= $this->elementEmpty('input', $attr); + + $ret .= $this->end('div'); + + return $ret; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Printer/HTMLDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/Printer/HTMLDefinition.php new file mode 100644 index 000000000..8a8f126b8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Printer/HTMLDefinition.php @@ -0,0 +1,272 @@ +<?php + +class HTMLPurifier_Printer_HTMLDefinition extends HTMLPurifier_Printer +{ + + /** + * Instance of HTMLPurifier_HTMLDefinition, for easy access + */ + protected $def; + + public function render($config) { + $ret = ''; + $this->config =& $config; + + $this->def = $config->getHTMLDefinition(); + + $ret .= $this->start('div', array('class' => 'HTMLPurifier_Printer')); + + $ret .= $this->renderDoctype(); + $ret .= $this->renderEnvironment(); + $ret .= $this->renderContentSets(); + $ret .= $this->renderInfo(); + + $ret .= $this->end('div'); + + return $ret; + } + + /** + * Renders the Doctype table + */ + protected function renderDoctype() { + $doctype = $this->def->doctype; + $ret = ''; + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Doctype'); + $ret .= $this->row('Name', $doctype->name); + $ret .= $this->row('XML', $doctype->xml ? 'Yes' : 'No'); + $ret .= $this->row('Default Modules', implode($doctype->modules, ', ')); + $ret .= $this->row('Default Tidy Modules', implode($doctype->tidyModules, ', ')); + $ret .= $this->end('table'); + return $ret; + } + + + /** + * Renders environment table, which is miscellaneous info + */ + protected function renderEnvironment() { + $def = $this->def; + + $ret = ''; + + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Environment'); + + $ret .= $this->row('Parent of fragment', $def->info_parent); + $ret .= $this->renderChildren($def->info_parent_def->child); + $ret .= $this->row('Block wrap name', $def->info_block_wrapper); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Global attributes'); + $ret .= $this->element('td', $this->listifyAttr($def->info_global_attr),0,0); + $ret .= $this->end('tr'); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Tag transforms'); + $list = array(); + foreach ($def->info_tag_transform as $old => $new) { + $new = $this->getClass($new, 'TagTransform_'); + $list[] = "<$old> with $new"; + } + $ret .= $this->element('td', $this->listify($list)); + $ret .= $this->end('tr'); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Pre-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->info_attr_transform_pre)); + $ret .= $this->end('tr'); + + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Post-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->info_attr_transform_post)); + $ret .= $this->end('tr'); + + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders the Content Sets table + */ + protected function renderContentSets() { + $ret = ''; + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Content Sets'); + foreach ($this->def->info_content_sets as $name => $lookup) { + $ret .= $this->heavyHeader($name); + $ret .= $this->start('tr'); + $ret .= $this->element('td', $this->listifyTagLookup($lookup)); + $ret .= $this->end('tr'); + } + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders the Elements ($info) table + */ + protected function renderInfo() { + $ret = ''; + $ret .= $this->start('table'); + $ret .= $this->element('caption', 'Elements ($info)'); + ksort($this->def->info); + $ret .= $this->heavyHeader('Allowed tags', 2); + $ret .= $this->start('tr'); + $ret .= $this->element('td', $this->listifyTagLookup($this->def->info), array('colspan' => 2)); + $ret .= $this->end('tr'); + foreach ($this->def->info as $name => $def) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', "<$name>", array('class'=>'heavy', 'colspan' => 2)); + $ret .= $this->end('tr'); + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Inline content'); + $ret .= $this->element('td', $def->descendants_are_inline ? 'Yes' : 'No'); + $ret .= $this->end('tr'); + if (!empty($def->excludes)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Excludes'); + $ret .= $this->element('td', $this->listifyTagLookup($def->excludes)); + $ret .= $this->end('tr'); + } + if (!empty($def->attr_transform_pre)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Pre-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->attr_transform_pre)); + $ret .= $this->end('tr'); + } + if (!empty($def->attr_transform_post)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Post-AttrTransform'); + $ret .= $this->element('td', $this->listifyObjectList($def->attr_transform_post)); + $ret .= $this->end('tr'); + } + if (!empty($def->auto_close)) { + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Auto closed by'); + $ret .= $this->element('td', $this->listifyTagLookup($def->auto_close)); + $ret .= $this->end('tr'); + } + $ret .= $this->start('tr'); + $ret .= $this->element('th', 'Allowed attributes'); + $ret .= $this->element('td',$this->listifyAttr($def->attr), array(), 0); + $ret .= $this->end('tr'); + + if (!empty($def->required_attr)) { + $ret .= $this->row('Required attributes', $this->listify($def->required_attr)); + } + + $ret .= $this->renderChildren($def->child); + } + $ret .= $this->end('table'); + return $ret; + } + + /** + * Renders a row describing the allowed children of an element + * @param $def HTMLPurifier_ChildDef of pertinent element + */ + protected function renderChildren($def) { + $context = new HTMLPurifier_Context(); + $ret = ''; + $ret .= $this->start('tr'); + $elements = array(); + $attr = array(); + if (isset($def->elements)) { + if ($def->type == 'strictblockquote') { + $def->validateChildren(array(), $this->config, $context); + } + $elements = $def->elements; + } + if ($def->type == 'chameleon') { + $attr['rowspan'] = 2; + } elseif ($def->type == 'empty') { + $elements = array(); + } elseif ($def->type == 'table') { + $elements = array_flip(array('col', 'caption', 'colgroup', 'thead', + 'tfoot', 'tbody', 'tr')); + } + $ret .= $this->element('th', 'Allowed children', $attr); + + if ($def->type == 'chameleon') { + + $ret .= $this->element('td', + '<em>Block</em>: ' . + $this->escape($this->listifyTagLookup($def->block->elements)),0,0); + $ret .= $this->end('tr'); + $ret .= $this->start('tr'); + $ret .= $this->element('td', + '<em>Inline</em>: ' . + $this->escape($this->listifyTagLookup($def->inline->elements)),0,0); + + } elseif ($def->type == 'custom') { + + $ret .= $this->element('td', '<em>'.ucfirst($def->type).'</em>: ' . + $def->dtd_regex); + + } else { + $ret .= $this->element('td', + '<em>'.ucfirst($def->type).'</em>: ' . + $this->escape($this->listifyTagLookup($elements)),0,0); + } + $ret .= $this->end('tr'); + return $ret; + } + + /** + * Listifies a tag lookup table. + * @param $array Tag lookup array in form of array('tagname' => true) + */ + protected function listifyTagLookup($array) { + ksort($array); + $list = array(); + foreach ($array as $name => $discard) { + if ($name !== '#PCDATA' && !isset($this->def->info[$name])) continue; + $list[] = $name; + } + return $this->listify($list); + } + + /** + * Listifies a list of objects by retrieving class names and internal state + * @param $array List of objects + * @todo Also add information about internal state + */ + protected function listifyObjectList($array) { + ksort($array); + $list = array(); + foreach ($array as $discard => $obj) { + $list[] = $this->getClass($obj, 'AttrTransform_'); + } + return $this->listify($list); + } + + /** + * Listifies a hash of attributes to AttrDef classes + * @param $array Array hash in form of array('attrname' => HTMLPurifier_AttrDef) + */ + protected function listifyAttr($array) { + ksort($array); + $list = array(); + foreach ($array as $name => $obj) { + if ($obj === false) continue; + $list[] = "$name = <i>" . $this->getClass($obj, 'AttrDef_') . '</i>'; + } + return $this->listify($list); + } + + /** + * Creates a heavy header row + */ + protected function heavyHeader($text, $num = 1) { + $ret = ''; + $ret .= $this->start('tr'); + $ret .= $this->element('th', $text, array('colspan' => $num, 'class' => 'heavy')); + $ret .= $this->end('tr'); + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/PropertyList.php b/lib/htmlpurifier/library/HTMLPurifier/PropertyList.php new file mode 100644 index 000000000..2b99fb7bc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/PropertyList.php @@ -0,0 +1,86 @@ +<?php + +/** + * Generic property list implementation + */ +class HTMLPurifier_PropertyList +{ + /** + * Internal data-structure for properties + */ + protected $data = array(); + + /** + * Parent plist + */ + protected $parent; + + protected $cache; + + public function __construct($parent = null) { + $this->parent = $parent; + } + + /** + * Recursively retrieves the value for a key + */ + public function get($name) { + if ($this->has($name)) return $this->data[$name]; + // possible performance bottleneck, convert to iterative if necessary + if ($this->parent) return $this->parent->get($name); + throw new HTMLPurifier_Exception("Key '$name' not found"); + } + + /** + * Sets the value of a key, for this plist + */ + public function set($name, $value) { + $this->data[$name] = $value; + } + + /** + * Returns true if a given key exists + */ + public function has($name) { + return array_key_exists($name, $this->data); + } + + /** + * Resets a value to the value of it's parent, usually the default. If + * no value is specified, the entire plist is reset. + */ + public function reset($name = null) { + if ($name == null) $this->data = array(); + else unset($this->data[$name]); + } + + /** + * Squashes this property list and all of its property lists into a single + * array, and returns the array. This value is cached by default. + * @param $force If true, ignores the cache and regenerates the array. + */ + public function squash($force = false) { + if ($this->cache !== null && !$force) return $this->cache; + if ($this->parent) { + return $this->cache = array_merge($this->parent->squash($force), $this->data); + } else { + return $this->cache = $this->data; + } + } + + /** + * Returns the parent plist. + */ + public function getParent() { + return $this->parent; + } + + /** + * Sets the parent plist. + */ + public function setParent($plist) { + $this->parent = $plist; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/PropertyListIterator.php b/lib/htmlpurifier/library/HTMLPurifier/PropertyListIterator.php new file mode 100644 index 000000000..8f250443e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/PropertyListIterator.php @@ -0,0 +1,32 @@ +<?php + +/** + * Property list iterator. Do not instantiate this class directly. + */ +class HTMLPurifier_PropertyListIterator extends FilterIterator +{ + + protected $l; + protected $filter; + + /** + * @param $data Array of data to iterate over + * @param $filter Optional prefix to only allow values of + */ + public function __construct(Iterator $iterator, $filter = null) { + parent::__construct($iterator); + $this->l = strlen($filter); + $this->filter = $filter; + } + + public function accept() { + $key = $this->getInnerIterator()->key(); + if( strncmp($key, $this->filter, $this->l) !== 0 ) { + return false; + } + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy.php new file mode 100644 index 000000000..246286521 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy.php @@ -0,0 +1,26 @@ +<?php + +/** + * Supertype for classes that define a strategy for modifying/purifying tokens. + * + * While HTMLPurifier's core purpose is fixing HTML into something proper, + * strategies provide plug points for extra configuration or even extra + * features, such as custom tags, custom parsing of text, etc. + */ + + +abstract class HTMLPurifier_Strategy +{ + + /** + * Executes the strategy on the tokens. + * + * @param $tokens Array of HTMLPurifier_Token objects to be operated on. + * @param $config Configuration options + * @returns Processed array of token objects. + */ + abstract public function execute($tokens, $config, $context); + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php new file mode 100644 index 000000000..92aefd33e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Composite.php @@ -0,0 +1,23 @@ +<?php + +/** + * Composite strategy that runs multiple strategies on tokens. + */ +abstract class HTMLPurifier_Strategy_Composite extends HTMLPurifier_Strategy +{ + + /** + * List of strategies to run tokens through. + */ + protected $strategies = array(); + + public function execute($tokens, $config, $context) { + foreach ($this->strategies as $strategy) { + $tokens = $strategy->execute($tokens, $config, $context); + } + return $tokens; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php new file mode 100644 index 000000000..d90e15860 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/Core.php @@ -0,0 +1,18 @@ +<?php + +/** + * Core strategy composed of the big four strategies. + */ +class HTMLPurifier_Strategy_Core extends HTMLPurifier_Strategy_Composite +{ + + public function __construct() { + $this->strategies[] = new HTMLPurifier_Strategy_RemoveForeignElements(); + $this->strategies[] = new HTMLPurifier_Strategy_MakeWellFormed(); + $this->strategies[] = new HTMLPurifier_Strategy_FixNesting(); + $this->strategies[] = new HTMLPurifier_Strategy_ValidateAttributes(); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php new file mode 100644 index 000000000..f81802391 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/FixNesting.php @@ -0,0 +1,328 @@ +<?php + +/** + * Takes a well formed list of tokens and fixes their nesting. + * + * HTML elements dictate which elements are allowed to be their children, + * for example, you can't have a p tag in a span tag. Other elements have + * much more rigorous definitions: tables, for instance, require a specific + * order for their elements. There are also constraints not expressible by + * document type definitions, such as the chameleon nature of ins/del + * tags and global child exclusions. + * + * The first major objective of this strategy is to iterate through all the + * nodes (not tokens) of the list of tokens and determine whether or not + * their children conform to the element's definition. If they do not, the + * child definition may optionally supply an amended list of elements that + * is valid or require that the entire node be deleted (and the previous + * node rescanned). + * + * The second objective is to ensure that explicitly excluded elements of + * an element do not appear in its children. Code that accomplishes this + * task is pervasive through the strategy, though the two are distinct tasks + * and could, theoretically, be seperated (although it's not recommended). + * + * @note Whether or not unrecognized children are silently dropped or + * translated into text depends on the child definitions. + * + * @todo Enable nodes to be bubbled out of the structure. + */ + +class HTMLPurifier_Strategy_FixNesting extends HTMLPurifier_Strategy +{ + + public function execute($tokens, $config, $context) { + //####################################################################// + // Pre-processing + + // get a copy of the HTML definition + $definition = $config->getHTMLDefinition(); + + // insert implicit "parent" node, will be removed at end. + // DEFINITION CALL + $parent_name = $definition->info_parent; + array_unshift($tokens, new HTMLPurifier_Token_Start($parent_name)); + $tokens[] = new HTMLPurifier_Token_End($parent_name); + + // setup the context variable 'IsInline', for chameleon processing + // is 'false' when we are not inline, 'true' when it must always + // be inline, and an integer when it is inline for a certain + // branch of the document tree + $is_inline = $definition->info_parent_def->descendants_are_inline; + $context->register('IsInline', $is_inline); + + // setup error collector + $e =& $context->get('ErrorCollector', true); + + //####################################################################// + // Loop initialization + + // stack that contains the indexes of all parents, + // $stack[count($stack)-1] being the current parent + $stack = array(); + + // stack that contains all elements that are excluded + // it is organized by parent elements, similar to $stack, + // but it is only populated when an element with exclusions is + // processed, i.e. there won't be empty exclusions. + $exclude_stack = array(); + + // variable that contains the start token while we are processing + // nodes. This enables error reporting to do its job + $start_token = false; + $context->register('CurrentToken', $start_token); + + //####################################################################// + // Loop + + // iterate through all start nodes. Determining the start node + // is complicated so it has been omitted from the loop construct + for ($i = 0, $size = count($tokens) ; $i < $size; ) { + + //################################################################// + // Gather information on children + + // child token accumulator + $child_tokens = array(); + + // scroll to the end of this node, report number, and collect + // all children + for ($j = $i, $depth = 0; ; $j++) { + if ($tokens[$j] instanceof HTMLPurifier_Token_Start) { + $depth++; + // skip token assignment on first iteration, this is the + // token we currently are on + if ($depth == 1) continue; + } elseif ($tokens[$j] instanceof HTMLPurifier_Token_End) { + $depth--; + // skip token assignment on last iteration, this is the + // end token of the token we're currently on + if ($depth == 0) break; + } + $child_tokens[] = $tokens[$j]; + } + + // $i is index of start token + // $j is index of end token + + $start_token = $tokens[$i]; // to make token available via CurrentToken + + //################################################################// + // Gather information on parent + + // calculate parent information + if ($count = count($stack)) { + $parent_index = $stack[$count-1]; + $parent_name = $tokens[$parent_index]->name; + if ($parent_index == 0) { + $parent_def = $definition->info_parent_def; + } else { + $parent_def = $definition->info[$parent_name]; + } + } else { + // processing as if the parent were the "root" node + // unknown info, it won't be used anyway, in the future, + // we may want to enforce one element only (this is + // necessary for HTML Purifier to clean entire documents + $parent_index = $parent_name = $parent_def = null; + } + + // calculate context + if ($is_inline === false) { + // check if conditions make it inline + if (!empty($parent_def) && $parent_def->descendants_are_inline) { + $is_inline = $count - 1; + } + } else { + // check if we're out of inline + if ($count === $is_inline) { + $is_inline = false; + } + } + + //################################################################// + // Determine whether element is explicitly excluded SGML-style + + // determine whether or not element is excluded by checking all + // parent exclusions. The array should not be very large, two + // elements at most. + $excluded = false; + if (!empty($exclude_stack)) { + foreach ($exclude_stack as $lookup) { + if (isset($lookup[$tokens[$i]->name])) { + $excluded = true; + // no need to continue processing + break; + } + } + } + + //################################################################// + // Perform child validation + + if ($excluded) { + // there is an exclusion, remove the entire node + $result = false; + $excludes = array(); // not used, but good to initialize anyway + } else { + // DEFINITION CALL + if ($i === 0) { + // special processing for the first node + $def = $definition->info_parent_def; + } else { + $def = $definition->info[$tokens[$i]->name]; + + } + + if (!empty($def->child)) { + // have DTD child def validate children + $result = $def->child->validateChildren( + $child_tokens, $config, $context); + } else { + // weird, no child definition, get rid of everything + $result = false; + } + + // determine whether or not this element has any exclusions + $excludes = $def->excludes; + } + + // $result is now a bool or array + + //################################################################// + // Process result by interpreting $result + + if ($result === true || $child_tokens === $result) { + // leave the node as is + + // register start token as a parental node start + $stack[] = $i; + + // register exclusions if there are any + if (!empty($excludes)) $exclude_stack[] = $excludes; + + // move cursor to next possible start node + $i++; + + } elseif($result === false) { + // remove entire node + + if ($e) { + if ($excluded) { + $e->send(E_ERROR, 'Strategy_FixNesting: Node excluded'); + } else { + $e->send(E_ERROR, 'Strategy_FixNesting: Node removed'); + } + } + + // calculate length of inner tokens and current tokens + $length = $j - $i + 1; + + // perform removal + array_splice($tokens, $i, $length); + + // update size + $size -= $length; + + // there is no start token to register, + // current node is now the next possible start node + // unless it turns out that we need to do a double-check + + // this is a rought heuristic that covers 100% of HTML's + // cases and 99% of all other cases. A child definition + // that would be tricked by this would be something like: + // ( | a b c) where it's all or nothing. Fortunately, + // our current implementation claims that that case would + // not allow empty, even if it did + if (!$parent_def->child->allow_empty) { + // we need to do a double-check + $i = $parent_index; + array_pop($stack); + } + + // PROJECTED OPTIMIZATION: Process all children elements before + // reprocessing parent node. + + } else { + // replace node with $result + + // calculate length of inner tokens + $length = $j - $i - 1; + + if ($e) { + if (empty($result) && $length) { + $e->send(E_ERROR, 'Strategy_FixNesting: Node contents removed'); + } else { + $e->send(E_WARNING, 'Strategy_FixNesting: Node reorganized'); + } + } + + // perform replacement + array_splice($tokens, $i + 1, $length, $result); + + // update size + $size -= $length; + $size += count($result); + + // register start token as a parental node start + $stack[] = $i; + + // register exclusions if there are any + if (!empty($excludes)) $exclude_stack[] = $excludes; + + // move cursor to next possible start node + $i++; + + } + + //################################################################// + // Scroll to next start node + + // We assume, at this point, that $i is the index of the token + // that is the first possible new start point for a node. + + // Test if the token indeed is a start tag, if not, move forward + // and test again. + $size = count($tokens); + while ($i < $size and !$tokens[$i] instanceof HTMLPurifier_Token_Start) { + if ($tokens[$i] instanceof HTMLPurifier_Token_End) { + // pop a token index off the stack if we ended a node + array_pop($stack); + // pop an exclusion lookup off exclusion stack if + // we ended node and that node had exclusions + if ($i == 0 || $i == $size - 1) { + // use specialized var if it's the super-parent + $s_excludes = $definition->info_parent_def->excludes; + } else { + $s_excludes = $definition->info[$tokens[$i]->name]->excludes; + } + if ($s_excludes) { + array_pop($exclude_stack); + } + } + $i++; + } + + } + + //####################################################################// + // Post-processing + + // remove implicit parent tokens at the beginning and end + array_shift($tokens); + array_pop($tokens); + + // remove context variables + $context->destroy('IsInline'); + $context->destroy('CurrentToken'); + + //####################################################################// + // Return + + return $tokens; + + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php new file mode 100644 index 000000000..c7aa1bb86 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/MakeWellFormed.php @@ -0,0 +1,532 @@ +<?php + +/** + * Takes tokens makes them well-formed (balance end tags, etc.) + * + * Specification of the armor attributes this strategy uses: + * + * - MakeWellFormed_TagClosedError: This armor field is used to + * suppress tag closed errors for certain tokens [TagClosedSuppress], + * in particular, if a tag was generated automatically by HTML + * Purifier, we may rely on our infrastructure to close it for us + * and shouldn't report an error to the user [TagClosedAuto]. + */ +class HTMLPurifier_Strategy_MakeWellFormed extends HTMLPurifier_Strategy +{ + + /** + * Array stream of tokens being processed. + */ + protected $tokens; + + /** + * Current index in $tokens. + */ + protected $t; + + /** + * Current nesting of elements. + */ + protected $stack; + + /** + * Injectors active in this stream processing. + */ + protected $injectors; + + /** + * Current instance of HTMLPurifier_Config. + */ + protected $config; + + /** + * Current instance of HTMLPurifier_Context. + */ + protected $context; + + public function execute($tokens, $config, $context) { + + $definition = $config->getHTMLDefinition(); + + // local variables + $generator = new HTMLPurifier_Generator($config, $context); + $escape_invalid_tags = $config->get('Core.EscapeInvalidTags'); + // used for autoclose early abortion + $global_parent_allowed_elements = array(); + if (isset($definition->info[$definition->info_parent])) { + // may be unset under testing circumstances + $global_parent_allowed_elements = $definition->info[$definition->info_parent]->child->getAllowedElements($config); + } + $e = $context->get('ErrorCollector', true); + $t = false; // token index + $i = false; // injector index + $token = false; // the current token + $reprocess = false; // whether or not to reprocess the same token + $stack = array(); + + // member variables + $this->stack =& $stack; + $this->t =& $t; + $this->tokens =& $tokens; + $this->config = $config; + $this->context = $context; + + // context variables + $context->register('CurrentNesting', $stack); + $context->register('InputIndex', $t); + $context->register('InputTokens', $tokens); + $context->register('CurrentToken', $token); + + // -- begin INJECTOR -- + + $this->injectors = array(); + + $injectors = $config->getBatch('AutoFormat'); + $def_injectors = $definition->info_injector; + $custom_injectors = $injectors['Custom']; + unset($injectors['Custom']); // special case + foreach ($injectors as $injector => $b) { + // XXX: Fix with a legitimate lookup table of enabled filters + if (strpos($injector, '.') !== false) continue; + $injector = "HTMLPurifier_Injector_$injector"; + if (!$b) continue; + $this->injectors[] = new $injector; + } + foreach ($def_injectors as $injector) { + // assumed to be objects + $this->injectors[] = $injector; + } + foreach ($custom_injectors as $injector) { + if (!$injector) continue; + if (is_string($injector)) { + $injector = "HTMLPurifier_Injector_$injector"; + $injector = new $injector; + } + $this->injectors[] = $injector; + } + + // give the injectors references to the definition and context + // variables for performance reasons + foreach ($this->injectors as $ix => $injector) { + $error = $injector->prepare($config, $context); + if (!$error) continue; + array_splice($this->injectors, $ix, 1); // rm the injector + trigger_error("Cannot enable {$injector->name} injector because $error is not allowed", E_USER_WARNING); + } + + // -- end INJECTOR -- + + // a note on reprocessing: + // In order to reduce code duplication, whenever some code needs + // to make HTML changes in order to make things "correct", the + // new HTML gets sent through the purifier, regardless of its + // status. This means that if we add a start token, because it + // was totally necessary, we don't have to update nesting; we just + // punt ($reprocess = true; continue;) and it does that for us. + + // isset is in loop because $tokens size changes during loop exec + for ( + $t = 0; + $t == 0 || isset($tokens[$t - 1]); + // only increment if we don't need to reprocess + $reprocess ? $reprocess = false : $t++ + ) { + + // check for a rewind + if (is_int($i) && $i >= 0) { + // possibility: disable rewinding if the current token has a + // rewind set on it already. This would offer protection from + // infinite loop, but might hinder some advanced rewinding. + $rewind_to = $this->injectors[$i]->getRewind(); + if (is_int($rewind_to) && $rewind_to < $t) { + if ($rewind_to < 0) $rewind_to = 0; + while ($t > $rewind_to) { + $t--; + $prev = $tokens[$t]; + // indicate that other injectors should not process this token, + // but we need to reprocess it + unset($prev->skip[$i]); + $prev->rewind = $i; + if ($prev instanceof HTMLPurifier_Token_Start) array_pop($this->stack); + elseif ($prev instanceof HTMLPurifier_Token_End) $this->stack[] = $prev->start; + } + } + $i = false; + } + + // handle case of document end + if (!isset($tokens[$t])) { + // kill processing if stack is empty + if (empty($this->stack)) break; + + // peek + $top_nesting = array_pop($this->stack); + $this->stack[] = $top_nesting; + + // send error [TagClosedSuppress] + if ($e && !isset($top_nesting->armor['MakeWellFormed_TagClosedError'])) { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag closed by document end', $top_nesting); + } + + // append, don't splice, since this is the end + $tokens[] = new HTMLPurifier_Token_End($top_nesting->name); + + // punt! + $reprocess = true; + continue; + } + + $token = $tokens[$t]; + + //echo '<br>'; printTokens($tokens, $t); printTokens($this->stack); + //flush(); + + // quick-check: if it's not a tag, no need to process + if (empty($token->is_tag)) { + if ($token instanceof HTMLPurifier_Token_Text) { + foreach ($this->injectors as $i => $injector) { + if (isset($token->skip[$i])) continue; + if ($token->rewind !== null && $token->rewind !== $i) continue; + $injector->handleText($token); + $this->processToken($token, $i); + $reprocess = true; + break; + } + } + // another possibility is a comment + continue; + } + + if (isset($definition->info[$token->name])) { + $type = $definition->info[$token->name]->child->type; + } else { + $type = false; // Type is unknown, treat accordingly + } + + // quick tag checks: anything that's *not* an end tag + $ok = false; + if ($type === 'empty' && $token instanceof HTMLPurifier_Token_Start) { + // claims to be a start tag but is empty + $token = new HTMLPurifier_Token_Empty($token->name, $token->attr, $token->line, $token->col, $token->armor); + $ok = true; + } elseif ($type && $type !== 'empty' && $token instanceof HTMLPurifier_Token_Empty) { + // claims to be empty but really is a start tag + $this->swap(new HTMLPurifier_Token_End($token->name)); + $this->insertBefore(new HTMLPurifier_Token_Start($token->name, $token->attr, $token->line, $token->col, $token->armor)); + // punt (since we had to modify the input stream in a non-trivial way) + $reprocess = true; + continue; + } elseif ($token instanceof HTMLPurifier_Token_Empty) { + // real empty token + $ok = true; + } elseif ($token instanceof HTMLPurifier_Token_Start) { + // start tag + + // ...unless they also have to close their parent + if (!empty($this->stack)) { + + // Performance note: you might think that it's rather + // inefficient, recalculating the autoclose information + // for every tag that a token closes (since when we + // do an autoclose, we push a new token into the + // stream and then /process/ that, before + // re-processing this token.) But this is + // necessary, because an injector can make an + // arbitrary transformations to the autoclosing + // tokens we introduce, so things may have changed + // in the meantime. Also, doing the inefficient thing is + // "easy" to reason about (for certain perverse definitions + // of "easy") + + $parent = array_pop($this->stack); + $this->stack[] = $parent; + + if (isset($definition->info[$parent->name])) { + $elements = $definition->info[$parent->name]->child->getAllowedElements($config); + $autoclose = !isset($elements[$token->name]); + } else { + $autoclose = false; + } + + if ($autoclose && $definition->info[$token->name]->wrap) { + // Check if an element can be wrapped by another + // element to make it valid in a context (for + // example, <ul><ul> needs a <li> in between) + $wrapname = $definition->info[$token->name]->wrap; + $wrapdef = $definition->info[$wrapname]; + $elements = $wrapdef->child->getAllowedElements($config); + $parent_elements = $definition->info[$parent->name]->child->getAllowedElements($config); + if (isset($elements[$token->name]) && isset($parent_elements[$wrapname])) { + $newtoken = new HTMLPurifier_Token_Start($wrapname); + $this->insertBefore($newtoken); + $reprocess = true; + continue; + } + } + + $carryover = false; + if ($autoclose && $definition->info[$parent->name]->formatting) { + $carryover = true; + } + + if ($autoclose) { + // check if this autoclose is doomed to fail + // (this rechecks $parent, which his harmless) + $autoclose_ok = isset($global_parent_allowed_elements[$token->name]); + if (!$autoclose_ok) { + foreach ($this->stack as $ancestor) { + $elements = $definition->info[$ancestor->name]->child->getAllowedElements($config); + if (isset($elements[$token->name])) { + $autoclose_ok = true; + break; + } + if ($definition->info[$token->name]->wrap) { + $wrapname = $definition->info[$token->name]->wrap; + $wrapdef = $definition->info[$wrapname]; + $wrap_elements = $wrapdef->child->getAllowedElements($config); + if (isset($wrap_elements[$token->name]) && isset($elements[$wrapname])) { + $autoclose_ok = true; + break; + } + } + } + } + if ($autoclose_ok) { + // errors need to be updated + $new_token = new HTMLPurifier_Token_End($parent->name); + $new_token->start = $parent; + if ($carryover) { + $element = clone $parent; + // [TagClosedAuto] + $element->armor['MakeWellFormed_TagClosedError'] = true; + $element->carryover = true; + $this->processToken(array($new_token, $token, $element)); + } else { + $this->insertBefore($new_token); + } + // [TagClosedSuppress] + if ($e && !isset($parent->armor['MakeWellFormed_TagClosedError'])) { + if (!$carryover) { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag auto closed', $parent); + } else { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag carryover', $parent); + } + } + } else { + $this->remove(); + } + $reprocess = true; + continue; + } + + } + $ok = true; + } + + if ($ok) { + foreach ($this->injectors as $i => $injector) { + if (isset($token->skip[$i])) continue; + if ($token->rewind !== null && $token->rewind !== $i) continue; + $injector->handleElement($token); + $this->processToken($token, $i); + $reprocess = true; + break; + } + if (!$reprocess) { + // ah, nothing interesting happened; do normal processing + $this->swap($token); + if ($token instanceof HTMLPurifier_Token_Start) { + $this->stack[] = $token; + } elseif ($token instanceof HTMLPurifier_Token_End) { + throw new HTMLPurifier_Exception('Improper handling of end tag in start code; possible error in MakeWellFormed'); + } + } + continue; + } + + // sanity check: we should be dealing with a closing tag + if (!$token instanceof HTMLPurifier_Token_End) { + throw new HTMLPurifier_Exception('Unaccounted for tag token in input stream, bug in HTML Purifier'); + } + + // make sure that we have something open + if (empty($this->stack)) { + if ($escape_invalid_tags) { + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Unnecessary end tag to text'); + $this->swap(new HTMLPurifier_Token_Text( + $generator->generateFromToken($token) + )); + } else { + $this->remove(); + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Unnecessary end tag removed'); + } + $reprocess = true; + continue; + } + + // first, check for the simplest case: everything closes neatly. + // Eventually, everything passes through here; if there are problems + // we modify the input stream accordingly and then punt, so that + // the tokens get processed again. + $current_parent = array_pop($this->stack); + if ($current_parent->name == $token->name) { + $token->start = $current_parent; + foreach ($this->injectors as $i => $injector) { + if (isset($token->skip[$i])) continue; + if ($token->rewind !== null && $token->rewind !== $i) continue; + $injector->handleEnd($token); + $this->processToken($token, $i); + $this->stack[] = $current_parent; + $reprocess = true; + break; + } + continue; + } + + // okay, so we're trying to close the wrong tag + + // undo the pop previous pop + $this->stack[] = $current_parent; + + // scroll back the entire nest, trying to find our tag. + // (feature could be to specify how far you'd like to go) + $size = count($this->stack); + // -2 because -1 is the last element, but we already checked that + $skipped_tags = false; + for ($j = $size - 2; $j >= 0; $j--) { + if ($this->stack[$j]->name == $token->name) { + $skipped_tags = array_slice($this->stack, $j); + break; + } + } + + // we didn't find the tag, so remove + if ($skipped_tags === false) { + if ($escape_invalid_tags) { + $this->swap(new HTMLPurifier_Token_Text( + $generator->generateFromToken($token) + )); + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Stray end tag to text'); + } else { + $this->remove(); + if ($e) $e->send(E_WARNING, 'Strategy_MakeWellFormed: Stray end tag removed'); + } + $reprocess = true; + continue; + } + + // do errors, in REVERSE $j order: a,b,c with </a></b></c> + $c = count($skipped_tags); + if ($e) { + for ($j = $c - 1; $j > 0; $j--) { + // notice we exclude $j == 0, i.e. the current ending tag, from + // the errors... [TagClosedSuppress] + if (!isset($skipped_tags[$j]->armor['MakeWellFormed_TagClosedError'])) { + $e->send(E_NOTICE, 'Strategy_MakeWellFormed: Tag closed by element end', $skipped_tags[$j]); + } + } + } + + // insert tags, in FORWARD $j order: c,b,a with </a></b></c> + $replace = array($token); + for ($j = 1; $j < $c; $j++) { + // ...as well as from the insertions + $new_token = new HTMLPurifier_Token_End($skipped_tags[$j]->name); + $new_token->start = $skipped_tags[$j]; + array_unshift($replace, $new_token); + if (isset($definition->info[$new_token->name]) && $definition->info[$new_token->name]->formatting) { + // [TagClosedAuto] + $element = clone $skipped_tags[$j]; + $element->carryover = true; + $element->armor['MakeWellFormed_TagClosedError'] = true; + $replace[] = $element; + } + } + $this->processToken($replace); + $reprocess = true; + continue; + } + + $context->destroy('CurrentNesting'); + $context->destroy('InputTokens'); + $context->destroy('InputIndex'); + $context->destroy('CurrentToken'); + + unset($this->injectors, $this->stack, $this->tokens, $this->t); + return $tokens; + } + + /** + * Processes arbitrary token values for complicated substitution patterns. + * In general: + * + * If $token is an array, it is a list of tokens to substitute for the + * current token. These tokens then get individually processed. If there + * is a leading integer in the list, that integer determines how many + * tokens from the stream should be removed. + * + * If $token is a regular token, it is swapped with the current token. + * + * If $token is false, the current token is deleted. + * + * If $token is an integer, that number of tokens (with the first token + * being the current one) will be deleted. + * + * @param $token Token substitution value + * @param $injector Injector that performed the substitution; default is if + * this is not an injector related operation. + */ + protected function processToken($token, $injector = -1) { + + // normalize forms of token + if (is_object($token)) $token = array(1, $token); + if (is_int($token)) $token = array($token); + if ($token === false) $token = array(1); + if (!is_array($token)) throw new HTMLPurifier_Exception('Invalid token type from injector'); + if (!is_int($token[0])) array_unshift($token, 1); + if ($token[0] === 0) throw new HTMLPurifier_Exception('Deleting zero tokens is not valid'); + + // $token is now an array with the following form: + // array(number nodes to delete, new node 1, new node 2, ...) + + $delete = array_shift($token); + $old = array_splice($this->tokens, $this->t, $delete, $token); + + if ($injector > -1) { + // determine appropriate skips + $oldskip = isset($old[0]) ? $old[0]->skip : array(); + foreach ($token as $object) { + $object->skip = $oldskip; + $object->skip[$injector] = true; + } + } + + } + + /** + * Inserts a token before the current token. Cursor now points to + * this token. You must reprocess after this. + */ + private function insertBefore($token) { + array_splice($this->tokens, $this->t, 0, array($token)); + } + + /** + * Removes current token. Cursor now points to new token occupying previously + * occupied space. You must reprocess after this. + */ + private function remove() { + array_splice($this->tokens, $this->t, 1); + } + + /** + * Swap current token with new token. Cursor points to new token (no + * change). You must reprocess after this. + */ + private function swap($token) { + $this->tokens[$this->t] = $token; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php new file mode 100644 index 000000000..bccaf14d3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/RemoveForeignElements.php @@ -0,0 +1,188 @@ +<?php + +/** + * Removes all unrecognized tags from the list of tokens. + * + * This strategy iterates through all the tokens and removes unrecognized + * tokens. If a token is not recognized but a TagTransform is defined for + * that element, the element will be transformed accordingly. + */ + +class HTMLPurifier_Strategy_RemoveForeignElements extends HTMLPurifier_Strategy +{ + + public function execute($tokens, $config, $context) { + $definition = $config->getHTMLDefinition(); + $generator = new HTMLPurifier_Generator($config, $context); + $result = array(); + + $escape_invalid_tags = $config->get('Core.EscapeInvalidTags'); + $remove_invalid_img = $config->get('Core.RemoveInvalidImg'); + + // currently only used to determine if comments should be kept + $trusted = $config->get('HTML.Trusted'); + $comment_lookup = $config->get('HTML.AllowedComments'); + $comment_regexp = $config->get('HTML.AllowedCommentsRegexp'); + $check_comments = $comment_lookup !== array() || $comment_regexp !== null; + + $remove_script_contents = $config->get('Core.RemoveScriptContents'); + $hidden_elements = $config->get('Core.HiddenElements'); + + // remove script contents compatibility + if ($remove_script_contents === true) { + $hidden_elements['script'] = true; + } elseif ($remove_script_contents === false && isset($hidden_elements['script'])) { + unset($hidden_elements['script']); + } + + $attr_validator = new HTMLPurifier_AttrValidator(); + + // removes tokens until it reaches a closing tag with its value + $remove_until = false; + + // converts comments into text tokens when this is equal to a tag name + $textify_comments = false; + + $token = false; + $context->register('CurrentToken', $token); + + $e = false; + if ($config->get('Core.CollectErrors')) { + $e =& $context->get('ErrorCollector'); + } + + foreach($tokens as $token) { + if ($remove_until) { + if (empty($token->is_tag) || $token->name !== $remove_until) { + continue; + } + } + if (!empty( $token->is_tag )) { + // DEFINITION CALL + + // before any processing, try to transform the element + if ( + isset($definition->info_tag_transform[$token->name]) + ) { + $original_name = $token->name; + // there is a transformation for this tag + // DEFINITION CALL + $token = $definition-> + info_tag_transform[$token->name]-> + transform($token, $config, $context); + if ($e) $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Tag transform', $original_name); + } + + if (isset($definition->info[$token->name])) { + + // mostly everything's good, but + // we need to make sure required attributes are in order + if ( + ($token instanceof HTMLPurifier_Token_Start || $token instanceof HTMLPurifier_Token_Empty) && + $definition->info[$token->name]->required_attr && + ($token->name != 'img' || $remove_invalid_img) // ensure config option still works + ) { + $attr_validator->validateToken($token, $config, $context); + $ok = true; + foreach ($definition->info[$token->name]->required_attr as $name) { + if (!isset($token->attr[$name])) { + $ok = false; + break; + } + } + if (!$ok) { + if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Missing required attribute', $name); + continue; + } + $token->armor['ValidateAttributes'] = true; + } + + if (isset($hidden_elements[$token->name]) && $token instanceof HTMLPurifier_Token_Start) { + $textify_comments = $token->name; + } elseif ($token->name === $textify_comments && $token instanceof HTMLPurifier_Token_End) { + $textify_comments = false; + } + + } elseif ($escape_invalid_tags) { + // invalid tag, generate HTML representation and insert in + if ($e) $e->send(E_WARNING, 'Strategy_RemoveForeignElements: Foreign element to text'); + $token = new HTMLPurifier_Token_Text( + $generator->generateFromToken($token) + ); + } else { + // check if we need to destroy all of the tag's children + // CAN BE GENERICIZED + if (isset($hidden_elements[$token->name])) { + if ($token instanceof HTMLPurifier_Token_Start) { + $remove_until = $token->name; + } elseif ($token instanceof HTMLPurifier_Token_Empty) { + // do nothing: we're still looking + } else { + $remove_until = false; + } + if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign meta element removed'); + } else { + if ($e) $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Foreign element removed'); + } + continue; + } + } elseif ($token instanceof HTMLPurifier_Token_Comment) { + // textify comments in script tags when they are allowed + if ($textify_comments !== false) { + $data = $token->data; + $token = new HTMLPurifier_Token_Text($data); + } elseif ($trusted || $check_comments) { + // always cleanup comments + $trailing_hyphen = false; + if ($e) { + // perform check whether or not there's a trailing hyphen + if (substr($token->data, -1) == '-') { + $trailing_hyphen = true; + } + } + $token->data = rtrim($token->data, '-'); + $found_double_hyphen = false; + while (strpos($token->data, '--') !== false) { + $found_double_hyphen = true; + $token->data = str_replace('--', '-', $token->data); + } + if ($trusted || !empty($comment_lookup[trim($token->data)]) || ($comment_regexp !== NULL && preg_match($comment_regexp, trim($token->data)))) { + // OK good + if ($e) { + if ($trailing_hyphen) { + $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Trailing hyphen in comment removed'); + } + if ($found_double_hyphen) { + $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Hyphens in comment collapsed'); + } + } + } else { + if ($e) { + $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed'); + } + continue; + } + } else { + // strip comments + if ($e) $e->send(E_NOTICE, 'Strategy_RemoveForeignElements: Comment removed'); + continue; + } + } elseif ($token instanceof HTMLPurifier_Token_Text) { + } else { + continue; + } + $result[] = $token; + } + if ($remove_until && $e) { + // we removed tokens until the end, throw error + $e->send(E_ERROR, 'Strategy_RemoveForeignElements: Token removed to end', $remove_until); + } + + $context->destroy('CurrentToken'); + + return $result; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php b/lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php new file mode 100644 index 000000000..c3328a9d4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Strategy/ValidateAttributes.php @@ -0,0 +1,39 @@ +<?php + +/** + * Validate all attributes in the tokens. + */ + +class HTMLPurifier_Strategy_ValidateAttributes extends HTMLPurifier_Strategy +{ + + public function execute($tokens, $config, $context) { + + // setup validator + $validator = new HTMLPurifier_AttrValidator(); + + $token = false; + $context->register('CurrentToken', $token); + + foreach ($tokens as $key => $token) { + + // only process tokens that have attributes, + // namely start and empty tags + if (!$token instanceof HTMLPurifier_Token_Start && !$token instanceof HTMLPurifier_Token_Empty) continue; + + // skip tokens that are armored + if (!empty($token->armor['ValidateAttributes'])) continue; + + // note that we have no facilities here for removing tokens + $validator->validateToken($token, $config, $context); + + $tokens[$key] = $token; // for PHP 4 + } + $context->destroy('CurrentToken'); + + return $tokens; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/StringHash.php b/lib/htmlpurifier/library/HTMLPurifier/StringHash.php new file mode 100644 index 000000000..62085c5c2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/StringHash.php @@ -0,0 +1,39 @@ +<?php + +/** + * This is in almost every respect equivalent to an array except + * that it keeps track of which keys were accessed. + * + * @warning For the sake of backwards compatibility with early versions + * of PHP 5, you must not use the $hash[$key] syntax; if you do + * our version of offsetGet is never called. + */ +class HTMLPurifier_StringHash extends ArrayObject +{ + protected $accessed = array(); + + /** + * Retrieves a value, and logs the access. + */ + public function offsetGet($index) { + $this->accessed[$index] = true; + return parent::offsetGet($index); + } + + /** + * Returns a lookup array of all array indexes that have been accessed. + * @return Array in form array($index => true). + */ + public function getAccessed() { + return $this->accessed; + } + + /** + * Resets the access array. + */ + public function resetAccessed() { + $this->accessed = array(); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/StringHashParser.php b/lib/htmlpurifier/library/HTMLPurifier/StringHashParser.php new file mode 100644 index 000000000..f3e70c712 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/StringHashParser.php @@ -0,0 +1,110 @@ +<?php + +/** + * Parses string hash files. File format is as such: + * + * DefaultKeyValue + * KEY: Value + * KEY2: Value2 + * --MULTILINE-KEY-- + * Multiline + * value. + * + * Which would output something similar to: + * + * array( + * 'ID' => 'DefaultKeyValue', + * 'KEY' => 'Value', + * 'KEY2' => 'Value2', + * 'MULTILINE-KEY' => "Multiline\nvalue.\n", + * ) + * + * We use this as an easy to use file-format for configuration schema + * files, but the class itself is usage agnostic. + * + * You can use ---- to forcibly terminate parsing of a single string-hash; + * this marker is used in multi string-hashes to delimit boundaries. + */ +class HTMLPurifier_StringHashParser +{ + + public $default = 'ID'; + + /** + * Parses a file that contains a single string-hash. + */ + public function parseFile($file) { + if (!file_exists($file)) return false; + $fh = fopen($file, 'r'); + if (!$fh) return false; + $ret = $this->parseHandle($fh); + fclose($fh); + return $ret; + } + + /** + * Parses a file that contains multiple string-hashes delimited by '----' + */ + public function parseMultiFile($file) { + if (!file_exists($file)) return false; + $ret = array(); + $fh = fopen($file, 'r'); + if (!$fh) return false; + while (!feof($fh)) { + $ret[] = $this->parseHandle($fh); + } + fclose($fh); + return $ret; + } + + /** + * Internal parser that acepts a file handle. + * @note While it's possible to simulate in-memory parsing by using + * custom stream wrappers, if such a use-case arises we should + * factor out the file handle into its own class. + * @param $fh File handle with pointer at start of valid string-hash + * block. + */ + protected function parseHandle($fh) { + $state = false; + $single = false; + $ret = array(); + do { + $line = fgets($fh); + if ($line === false) break; + $line = rtrim($line, "\n\r"); + if (!$state && $line === '') continue; + if ($line === '----') break; + if (strncmp('--#', $line, 3) === 0) { + // Comment + continue; + } elseif (strncmp('--', $line, 2) === 0) { + // Multiline declaration + $state = trim($line, '- '); + if (!isset($ret[$state])) $ret[$state] = ''; + continue; + } elseif (!$state) { + $single = true; + if (strpos($line, ':') !== false) { + // Single-line declaration + list($state, $line) = explode(':', $line, 2); + $line = trim($line); + } else { + // Use default declaration + $state = $this->default; + } + } + if ($single) { + $ret[$state] = $line; + $single = false; + $state = false; + } else { + $ret[$state] .= "$line\n"; + } + } while (!feof($fh)); + return $ret; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TagTransform.php b/lib/htmlpurifier/library/HTMLPurifier/TagTransform.php new file mode 100644 index 000000000..210a44721 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TagTransform.php @@ -0,0 +1,36 @@ +<?php + +/** + * Defines a mutation of an obsolete tag into a valid tag. + */ +abstract class HTMLPurifier_TagTransform +{ + + /** + * Tag name to transform the tag to. + */ + public $transform_to; + + /** + * Transforms the obsolete tag into the valid tag. + * @param $tag Tag to be transformed. + * @param $config Mandatory HTMLPurifier_Config object + * @param $context Mandatory HTMLPurifier_Context object + */ + abstract public function transform($tag, $config, $context); + + /** + * Prepends CSS properties to the style attribute, creating the + * attribute if it doesn't exist. + * @warning Copied over from AttrTransform, be sure to keep in sync + * @param $attr Attribute array to process (passed by reference) + * @param $css CSS to prepend + */ + protected function prependCSS(&$attr, $css) { + $attr['style'] = isset($attr['style']) ? $attr['style'] : ''; + $attr['style'] = $css . $attr['style']; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Font.php b/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Font.php new file mode 100644 index 000000000..9db2db795 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Font.php @@ -0,0 +1,98 @@ +<?php + +/** + * Transforms FONT tags to the proper form (SPAN with CSS styling) + * + * This transformation takes the three proprietary attributes of FONT and + * transforms them into their corresponding CSS attributes. These are color, + * face, and size. + * + * @note Size is an interesting case because it doesn't map cleanly to CSS. + * Thanks to + * http://style.cleverchimp.com/font_size_intervals/altintervals.html + * for reasonable mappings. + * @warning This doesn't work completely correctly; specifically, this + * TagTransform operates before well-formedness is enforced, so + * the "active formatting elements" algorithm doesn't get applied. + */ +class HTMLPurifier_TagTransform_Font extends HTMLPurifier_TagTransform +{ + + public $transform_to = 'span'; + + protected $_size_lookup = array( + '0' => 'xx-small', + '1' => 'xx-small', + '2' => 'small', + '3' => 'medium', + '4' => 'large', + '5' => 'x-large', + '6' => 'xx-large', + '7' => '300%', + '-1' => 'smaller', + '-2' => '60%', + '+1' => 'larger', + '+2' => '150%', + '+3' => '200%', + '+4' => '300%' + ); + + public function transform($tag, $config, $context) { + + if ($tag instanceof HTMLPurifier_Token_End) { + $new_tag = clone $tag; + $new_tag->name = $this->transform_to; + return $new_tag; + } + + $attr = $tag->attr; + $prepend_style = ''; + + // handle color transform + if (isset($attr['color'])) { + $prepend_style .= 'color:' . $attr['color'] . ';'; + unset($attr['color']); + } + + // handle face transform + if (isset($attr['face'])) { + $prepend_style .= 'font-family:' . $attr['face'] . ';'; + unset($attr['face']); + } + + // handle size transform + if (isset($attr['size'])) { + // normalize large numbers + if ($attr['size'] !== '') { + if ($attr['size']{0} == '+' || $attr['size']{0} == '-') { + $size = (int) $attr['size']; + if ($size < -2) $attr['size'] = '-2'; + if ($size > 4) $attr['size'] = '+4'; + } else { + $size = (int) $attr['size']; + if ($size > 7) $attr['size'] = '7'; + } + } + if (isset($this->_size_lookup[$attr['size']])) { + $prepend_style .= 'font-size:' . + $this->_size_lookup[$attr['size']] . ';'; + } + unset($attr['size']); + } + + if ($prepend_style) { + $attr['style'] = isset($attr['style']) ? + $prepend_style . $attr['style'] : + $prepend_style; + } + + $new_tag = clone $tag; + $new_tag->name = $this->transform_to; + $new_tag->attr = $attr; + + return $new_tag; + + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Simple.php b/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Simple.php new file mode 100644 index 000000000..0e36130f2 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TagTransform/Simple.php @@ -0,0 +1,35 @@ +<?php + +/** + * Simple transformation, just change tag name to something else, + * and possibly add some styling. This will cover most of the deprecated + * tag cases. + */ +class HTMLPurifier_TagTransform_Simple extends HTMLPurifier_TagTransform +{ + + protected $style; + + /** + * @param $transform_to Tag name to transform to. + * @param $style CSS style to add to the tag + */ + public function __construct($transform_to, $style = null) { + $this->transform_to = $transform_to; + $this->style = $style; + } + + public function transform($tag, $config, $context) { + $new_tag = clone $tag; + $new_tag->name = $this->transform_to; + if (!is_null($this->style) && + ($new_tag instanceof HTMLPurifier_Token_Start || $new_tag instanceof HTMLPurifier_Token_Empty) + ) { + $this->prependCSS($new_tag->attr, $this->style); + } + return $new_tag; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token.php b/lib/htmlpurifier/library/HTMLPurifier/Token.php new file mode 100644 index 000000000..7900e6cb1 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token.php @@ -0,0 +1,57 @@ +<?php + +/** + * Abstract base token class that all others inherit from. + */ +class HTMLPurifier_Token { + public $line; /**< Line number node was on in source document. Null if unknown. */ + public $col; /**< Column of line node was on in source document. Null if unknown. */ + + /** + * Lookup array of processing that this token is exempt from. + * Currently, valid values are "ValidateAttributes" and + * "MakeWellFormed_TagClosedError" + */ + public $armor = array(); + + /** + * Used during MakeWellFormed. + */ + public $skip; + public $rewind; + public $carryover; + + public function __get($n) { + if ($n === 'type') { + trigger_error('Deprecated type property called; use instanceof', E_USER_NOTICE); + switch (get_class($this)) { + case 'HTMLPurifier_Token_Start': return 'start'; + case 'HTMLPurifier_Token_Empty': return 'empty'; + case 'HTMLPurifier_Token_End': return 'end'; + case 'HTMLPurifier_Token_Text': return 'text'; + case 'HTMLPurifier_Token_Comment': return 'comment'; + default: return null; + } + } + } + + /** + * Sets the position of the token in the source document. + */ + public function position($l = null, $c = null) { + $this->line = $l; + $this->col = $c; + } + + /** + * Convenience function for DirectLex settings line/col position. + */ + public function rawPosition($l, $c) { + if ($c === -1) $l++; + $this->line = $l; + $this->col = $c; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php new file mode 100644 index 000000000..dc6bdcabb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Comment.php @@ -0,0 +1,22 @@ +<?php + +/** + * Concrete comment token class. Generally will be ignored. + */ +class HTMLPurifier_Token_Comment extends HTMLPurifier_Token +{ + public $data; /**< Character data within comment. */ + public $is_whitespace = true; + /** + * Transparent constructor. + * + * @param $data String comment data. + */ + public function __construct($data, $line = null, $col = null) { + $this->data = $data; + $this->line = $line; + $this->col = $col; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php new file mode 100644 index 000000000..2a82b47ad --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Empty.php @@ -0,0 +1,11 @@ +<?php + +/** + * Concrete empty token class. + */ +class HTMLPurifier_Token_Empty extends HTMLPurifier_Token_Tag +{ + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/End.php b/lib/htmlpurifier/library/HTMLPurifier/Token/End.php new file mode 100644 index 000000000..353e79daf --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/End.php @@ -0,0 +1,19 @@ +<?php + +/** + * Concrete end token class. + * + * @warning This class accepts attributes even though end tags cannot. This + * is for optimization reasons, as under normal circumstances, the Lexers + * do not pass attributes. + */ +class HTMLPurifier_Token_End extends HTMLPurifier_Token_Tag +{ + /** + * Token that started this node. Added by MakeWellFormed. Please + * do not edit this! + */ + public $start; +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Start.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Start.php new file mode 100644 index 000000000..e0e14fc62 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Start.php @@ -0,0 +1,11 @@ +<?php + +/** + * Concrete start token class. + */ +class HTMLPurifier_Token_Start extends HTMLPurifier_Token_Tag +{ + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php new file mode 100644 index 000000000..f4d8f640e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Tag.php @@ -0,0 +1,57 @@ +<?php + +/** + * Abstract class of a tag token (start, end or empty), and its behavior. + */ +class HTMLPurifier_Token_Tag extends HTMLPurifier_Token +{ + /** + * Static bool marker that indicates the class is a tag. + * + * This allows us to check objects with <tt>!empty($obj->is_tag)</tt> + * without having to use a function call <tt>is_a()</tt>. + */ + public $is_tag = true; + + /** + * The lower-case name of the tag, like 'a', 'b' or 'blockquote'. + * + * @note Strictly speaking, XML tags are case sensitive, so we shouldn't + * be lower-casing them, but these tokens cater to HTML tags, which are + * insensitive. + */ + public $name; + + /** + * Associative array of the tag's attributes. + */ + public $attr = array(); + + /** + * Non-overloaded constructor, which lower-cases passed tag name. + * + * @param $name String name. + * @param $attr Associative array of attributes. + */ + public function __construct($name, $attr = array(), $line = null, $col = null, $armor = array()) { + $this->name = ctype_lower($name) ? $name : strtolower($name); + foreach ($attr as $key => $value) { + // normalization only necessary when key is not lowercase + if (!ctype_lower($key)) { + $new_key = strtolower($key); + if (!isset($attr[$new_key])) { + $attr[$new_key] = $attr[$key]; + } + if ($new_key !== $key) { + unset($attr[$key]); + } + } + } + $this->attr = $attr; + $this->line = $line; + $this->col = $col; + $this->armor = $armor; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php b/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php new file mode 100644 index 000000000..82efd823d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/Token/Text.php @@ -0,0 +1,33 @@ +<?php + +/** + * Concrete text token class. + * + * Text tokens comprise of regular parsed character data (PCDATA) and raw + * character data (from the CDATA sections). Internally, their + * data is parsed with all entities expanded. Surprisingly, the text token + * does have a "tag name" called #PCDATA, which is how the DTD represents it + * in permissible child nodes. + */ +class HTMLPurifier_Token_Text extends HTMLPurifier_Token +{ + + public $name = '#PCDATA'; /**< PCDATA tag name compatible with DTD. */ + public $data; /**< Parsed character data of text. */ + public $is_whitespace; /**< Bool indicating if node is whitespace. */ + + /** + * Constructor, accepts data and determines if it is whitespace. + * + * @param $data String parsed character data. + */ + public function __construct($data, $line = null, $col = null) { + $this->data = $data; + $this->is_whitespace = ctype_space($data); + $this->line = $line; + $this->col = $col; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/TokenFactory.php b/lib/htmlpurifier/library/HTMLPurifier/TokenFactory.php new file mode 100644 index 000000000..7cf48fb41 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/TokenFactory.php @@ -0,0 +1,94 @@ +<?php + +/** + * Factory for token generation. + * + * @note Doing some benchmarking indicates that the new operator is much + * slower than the clone operator (even discounting the cost of the + * constructor). This class is for that optimization. + * Other then that, there's not much point as we don't + * maintain parallel HTMLPurifier_Token hierarchies (the main reason why + * you'd want to use an abstract factory). + * @todo Port DirectLex to use this + */ +class HTMLPurifier_TokenFactory +{ + + /** + * Prototypes that will be cloned. + * @private + */ + // p stands for prototype + private $p_start, $p_end, $p_empty, $p_text, $p_comment; + + /** + * Generates blank prototypes for cloning. + */ + public function __construct() { + $this->p_start = new HTMLPurifier_Token_Start('', array()); + $this->p_end = new HTMLPurifier_Token_End(''); + $this->p_empty = new HTMLPurifier_Token_Empty('', array()); + $this->p_text = new HTMLPurifier_Token_Text(''); + $this->p_comment= new HTMLPurifier_Token_Comment(''); + } + + /** + * Creates a HTMLPurifier_Token_Start. + * @param $name Tag name + * @param $attr Associative array of attributes + * @return Generated HTMLPurifier_Token_Start + */ + public function createStart($name, $attr = array()) { + $p = clone $this->p_start; + $p->__construct($name, $attr); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_End. + * @param $name Tag name + * @return Generated HTMLPurifier_Token_End + */ + public function createEnd($name) { + $p = clone $this->p_end; + $p->__construct($name); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_Empty. + * @param $name Tag name + * @param $attr Associative array of attributes + * @return Generated HTMLPurifier_Token_Empty + */ + public function createEmpty($name, $attr = array()) { + $p = clone $this->p_empty; + $p->__construct($name, $attr); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_Text. + * @param $data Data of text token + * @return Generated HTMLPurifier_Token_Text + */ + public function createText($data) { + $p = clone $this->p_text; + $p->__construct($data); + return $p; + } + + /** + * Creates a HTMLPurifier_Token_Comment. + * @param $data Data of comment token + * @return Generated HTMLPurifier_Token_Comment + */ + public function createComment($data) { + $p = clone $this->p_comment; + $p->__construct($data); + return $p; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URI.php b/lib/htmlpurifier/library/HTMLPurifier/URI.php new file mode 100644 index 000000000..f158ef5e3 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URI.php @@ -0,0 +1,242 @@ +<?php + +/** + * HTML Purifier's internal representation of a URI. + * @note + * Internal data-structures are completely escaped. If the data needs + * to be used in a non-URI context (which is very unlikely), be sure + * to decode it first. The URI may not necessarily be well-formed until + * validate() is called. + */ +class HTMLPurifier_URI +{ + + public $scheme, $userinfo, $host, $port, $path, $query, $fragment; + + /** + * @note Automatically normalizes scheme and port + */ + public function __construct($scheme, $userinfo, $host, $port, $path, $query, $fragment) { + $this->scheme = is_null($scheme) || ctype_lower($scheme) ? $scheme : strtolower($scheme); + $this->userinfo = $userinfo; + $this->host = $host; + $this->port = is_null($port) ? $port : (int) $port; + $this->path = $path; + $this->query = $query; + $this->fragment = $fragment; + } + + /** + * Retrieves a scheme object corresponding to the URI's scheme/default + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return Scheme object appropriate for validating this URI + */ + public function getSchemeObj($config, $context) { + $registry = HTMLPurifier_URISchemeRegistry::instance(); + if ($this->scheme !== null) { + $scheme_obj = $registry->getScheme($this->scheme, $config, $context); + if (!$scheme_obj) return false; // invalid scheme, clean it out + } else { + // no scheme: retrieve the default one + $def = $config->getDefinition('URI'); + $scheme_obj = $def->getDefaultScheme($config, $context); + if (!$scheme_obj) { + // something funky happened to the default scheme object + trigger_error( + 'Default scheme object "' . $def->defaultScheme . '" was not readable', + E_USER_WARNING + ); + return false; + } + } + return $scheme_obj; + } + + /** + * Generic validation method applicable for all schemes. May modify + * this URI in order to get it into a compliant form. + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return True if validation/filtering succeeds, false if failure + */ + public function validate($config, $context) { + + // ABNF definitions from RFC 3986 + $chars_sub_delims = '!$&\'()*+,;='; + $chars_gen_delims = ':/?#[]@'; + $chars_pchar = $chars_sub_delims . ':@'; + + // validate host + if (!is_null($this->host)) { + $host_def = new HTMLPurifier_AttrDef_URI_Host(); + $this->host = $host_def->validate($this->host, $config, $context); + if ($this->host === false) $this->host = null; + } + + // validate scheme + // NOTE: It's not appropriate to check whether or not this + // scheme is in our registry, since a URIFilter may convert a + // URI that we don't allow into one we do. So instead, we just + // check if the scheme can be dropped because there is no host + // and it is our default scheme. + if (!is_null($this->scheme) && is_null($this->host) || $this->host === '') { + // support for relative paths is pretty abysmal when the + // scheme is present, so axe it when possible + $def = $config->getDefinition('URI'); + if ($def->defaultScheme === $this->scheme) { + $this->scheme = null; + } + } + + // validate username + if (!is_null($this->userinfo)) { + $encoder = new HTMLPurifier_PercentEncoder($chars_sub_delims . ':'); + $this->userinfo = $encoder->encode($this->userinfo); + } + + // validate port + if (!is_null($this->port)) { + if ($this->port < 1 || $this->port > 65535) $this->port = null; + } + + // validate path + $path_parts = array(); + $segments_encoder = new HTMLPurifier_PercentEncoder($chars_pchar . '/'); + if (!is_null($this->host)) { // this catches $this->host === '' + // path-abempty (hier and relative) + // http://www.example.com/my/path + // //www.example.com/my/path (looks odd, but works, and + // recognized by most browsers) + // (this set is valid or invalid on a scheme by scheme + // basis, so we'll deal with it later) + // file:///my/path + // ///my/path + $this->path = $segments_encoder->encode($this->path); + } elseif ($this->path !== '') { + if ($this->path[0] === '/') { + // path-absolute (hier and relative) + // http:/my/path + // /my/path + if (strlen($this->path) >= 2 && $this->path[1] === '/') { + // This could happen if both the host gets stripped + // out + // http://my/path + // //my/path + $this->path = ''; + } else { + $this->path = $segments_encoder->encode($this->path); + } + } elseif (!is_null($this->scheme)) { + // path-rootless (hier) + // http:my/path + // Short circuit evaluation means we don't need to check nz + $this->path = $segments_encoder->encode($this->path); + } else { + // path-noscheme (relative) + // my/path + // (once again, not checking nz) + $segment_nc_encoder = new HTMLPurifier_PercentEncoder($chars_sub_delims . '@'); + $c = strpos($this->path, '/'); + if ($c !== false) { + $this->path = + $segment_nc_encoder->encode(substr($this->path, 0, $c)) . + $segments_encoder->encode(substr($this->path, $c)); + } else { + $this->path = $segment_nc_encoder->encode($this->path); + } + } + } else { + // path-empty (hier and relative) + $this->path = ''; // just to be safe + } + + // qf = query and fragment + $qf_encoder = new HTMLPurifier_PercentEncoder($chars_pchar . '/?'); + + if (!is_null($this->query)) { + $this->query = $qf_encoder->encode($this->query); + } + + if (!is_null($this->fragment)) { + $this->fragment = $qf_encoder->encode($this->fragment); + } + + return true; + + } + + /** + * Convert URI back to string + * @return String URI appropriate for output + */ + public function toString() { + // reconstruct authority + $authority = null; + // there is a rendering difference between a null authority + // (http:foo-bar) and an empty string authority + // (http:///foo-bar). + if (!is_null($this->host)) { + $authority = ''; + if(!is_null($this->userinfo)) $authority .= $this->userinfo . '@'; + $authority .= $this->host; + if(!is_null($this->port)) $authority .= ':' . $this->port; + } + + // Reconstruct the result + // One might wonder about parsing quirks from browsers after + // this reconstruction. Unfortunately, parsing behavior depends + // on what *scheme* was employed (file:///foo is handled *very* + // differently than http:///foo), so unfortunately we have to + // defer to the schemes to do the right thing. + $result = ''; + if (!is_null($this->scheme)) $result .= $this->scheme . ':'; + if (!is_null($authority)) $result .= '//' . $authority; + $result .= $this->path; + if (!is_null($this->query)) $result .= '?' . $this->query; + if (!is_null($this->fragment)) $result .= '#' . $this->fragment; + + return $result; + } + + /** + * Returns true if this URL might be considered a 'local' URL given + * the current context. This is true when the host is null, or + * when it matches the host supplied to the configuration. + * + * Note that this does not do any scheme checking, so it is mostly + * only appropriate for metadata that doesn't care about protocol + * security. isBenign is probably what you actually want. + */ + public function isLocal($config, $context) { + if ($this->host === null) return true; + $uri_def = $config->getDefinition('URI'); + if ($uri_def->host === $this->host) return true; + return false; + } + + /** + * Returns true if this URL should be considered a 'benign' URL, + * that is: + * + * - It is a local URL (isLocal), and + * - It has a equal or better level of security + */ + public function isBenign($config, $context) { + if (!$this->isLocal($config, $context)) return false; + + $scheme_obj = $this->getSchemeObj($config, $context); + if (!$scheme_obj) return false; // conservative approach + + $current_scheme_obj = $config->getDefinition('URI')->getDefaultScheme($config, $context); + if ($current_scheme_obj->secure) { + if (!$scheme_obj->secure) { + return false; + } + } + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIDefinition.php b/lib/htmlpurifier/library/HTMLPurifier/URIDefinition.php new file mode 100644 index 000000000..40e57bb7d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIDefinition.php @@ -0,0 +1,102 @@ +<?php + +class HTMLPurifier_URIDefinition extends HTMLPurifier_Definition +{ + + public $type = 'URI'; + protected $filters = array(); + protected $postFilters = array(); + protected $registeredFilters = array(); + + /** + * HTMLPurifier_URI object of the base specified at %URI.Base + */ + public $base; + + /** + * String host to consider "home" base, derived off of $base + */ + public $host; + + /** + * Name of default scheme based on %URI.DefaultScheme and %URI.Base + */ + public $defaultScheme; + + public function __construct() { + $this->registerFilter(new HTMLPurifier_URIFilter_DisableExternal()); + $this->registerFilter(new HTMLPurifier_URIFilter_DisableExternalResources()); + $this->registerFilter(new HTMLPurifier_URIFilter_HostBlacklist()); + $this->registerFilter(new HTMLPurifier_URIFilter_SafeIframe()); + $this->registerFilter(new HTMLPurifier_URIFilter_MakeAbsolute()); + $this->registerFilter(new HTMLPurifier_URIFilter_Munge()); + } + + public function registerFilter($filter) { + $this->registeredFilters[$filter->name] = $filter; + } + + public function addFilter($filter, $config) { + $r = $filter->prepare($config); + if ($r === false) return; // null is ok, for backwards compat + if ($filter->post) { + $this->postFilters[$filter->name] = $filter; + } else { + $this->filters[$filter->name] = $filter; + } + } + + protected function doSetup($config) { + $this->setupMemberVariables($config); + $this->setupFilters($config); + } + + protected function setupFilters($config) { + foreach ($this->registeredFilters as $name => $filter) { + if ($filter->always_load) { + $this->addFilter($filter, $config); + } else { + $conf = $config->get('URI.' . $name); + if ($conf !== false && $conf !== null) { + $this->addFilter($filter, $config); + } + } + } + unset($this->registeredFilters); + } + + protected function setupMemberVariables($config) { + $this->host = $config->get('URI.Host'); + $base_uri = $config->get('URI.Base'); + if (!is_null($base_uri)) { + $parser = new HTMLPurifier_URIParser(); + $this->base = $parser->parse($base_uri); + $this->defaultScheme = $this->base->scheme; + if (is_null($this->host)) $this->host = $this->base->host; + } + if (is_null($this->defaultScheme)) $this->defaultScheme = $config->get('URI.DefaultScheme'); + } + + public function getDefaultScheme($config, $context) { + return HTMLPurifier_URISchemeRegistry::instance()->getScheme($this->defaultScheme, $config, $context); + } + + public function filter(&$uri, $config, $context) { + foreach ($this->filters as $name => $f) { + $result = $f->filter($uri, $config, $context); + if (!$result) return false; + } + return true; + } + + public function postFilter(&$uri, $config, $context) { + foreach ($this->postFilters as $name => $f) { + $result = $f->filter($uri, $config, $context); + if (!$result) return false; + } + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter.php new file mode 100644 index 000000000..6a1b0b08e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter.php @@ -0,0 +1,67 @@ +<?php + +/** + * Chainable filters for custom URI processing. + * + * These filters can perform custom actions on a URI filter object, + * including transformation or blacklisting. A filter named Foo + * must have a corresponding configuration directive %URI.Foo, + * unless always_load is specified to be true. + * + * The following contexts may be available while URIFilters are being + * processed: + * + * - EmbeddedURI: true if URI is an embedded resource that will + * be loaded automatically on page load + * - CurrentToken: a reference to the token that is currently + * being processed + * - CurrentAttr: the name of the attribute that is currently being + * processed + * - CurrentCSSProperty: the name of the CSS property that is + * currently being processed (if applicable) + * + * @warning This filter is called before scheme object validation occurs. + * Make sure, if you require a specific scheme object, you + * you check that it exists. This allows filters to convert + * proprietary URI schemes into regular ones. + */ +abstract class HTMLPurifier_URIFilter +{ + + /** + * Unique identifier of filter + */ + public $name; + + /** + * True if this filter should be run after scheme validation. + */ + public $post = false; + + /** + * True if this filter should always be loaded (this permits + * a filter to be named Foo without the corresponding %URI.Foo + * directive existing.) + */ + public $always_load = false; + + /** + * Performs initialization for the filter. If the filter returns + * false, this means that it shouldn't be considered active. + */ + public function prepare($config) {return true;} + + /** + * Filter a URI object + * @param $uri Reference to URI object variable + * @param $config Instance of HTMLPurifier_Config + * @param $context Instance of HTMLPurifier_Context + * @return bool Whether or not to continue processing: false indicates + * URL is no good, true indicates continue processing. Note that + * all changes are committed directly on the URI object + */ + abstract public function filter(&$uri, $config, $context); + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternal.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternal.php new file mode 100644 index 000000000..d8a39a501 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternal.php @@ -0,0 +1,23 @@ +<?php + +class HTMLPurifier_URIFilter_DisableExternal extends HTMLPurifier_URIFilter +{ + public $name = 'DisableExternal'; + protected $ourHostParts = false; + public function prepare($config) { + $our_host = $config->getDefinition('URI')->host; + if ($our_host !== null) $this->ourHostParts = array_reverse(explode('.', $our_host)); + } + public function filter(&$uri, $config, $context) { + if (is_null($uri->host)) return true; + if ($this->ourHostParts === false) return false; + $host_parts = array_reverse(explode('.', $uri->host)); + foreach ($this->ourHostParts as $i => $x) { + if (!isset($host_parts[$i])) return false; + if ($host_parts[$i] != $this->ourHostParts[$i]) return false; + } + return true; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternalResources.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternalResources.php new file mode 100644 index 000000000..881abc43c --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableExternalResources.php @@ -0,0 +1,12 @@ +<?php + +class HTMLPurifier_URIFilter_DisableExternalResources extends HTMLPurifier_URIFilter_DisableExternal +{ + public $name = 'DisableExternalResources'; + public function filter(&$uri, $config, $context) { + if (!$context->get('EmbeddedURI', true)) return true; + return parent::filter($uri, $config, $context); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableResources.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableResources.php new file mode 100644 index 000000000..67538c7bb --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/DisableResources.php @@ -0,0 +1,11 @@ +<?php + +class HTMLPurifier_URIFilter_DisableResources extends HTMLPurifier_URIFilter +{ + public $name = 'DisableResources'; + public function filter(&$uri, $config, $context) { + return !$context->get('EmbeddedURI', true); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/HostBlacklist.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/HostBlacklist.php new file mode 100644 index 000000000..55fde3bf4 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/HostBlacklist.php @@ -0,0 +1,25 @@ +<?php + +// It's not clear to me whether or not Punycode means that hostnames +// do not have canonical forms anymore. As far as I can tell, it's +// not a problem (punycoding should be identity when no Unicode +// points are involved), but I'm not 100% sure +class HTMLPurifier_URIFilter_HostBlacklist extends HTMLPurifier_URIFilter +{ + public $name = 'HostBlacklist'; + protected $blacklist = array(); + public function prepare($config) { + $this->blacklist = $config->get('URI.HostBlacklist'); + return true; + } + public function filter(&$uri, $config, $context) { + foreach($this->blacklist as $blacklisted_host_fragment) { + if (strpos($uri->host, $blacklisted_host_fragment) !== false) { + return false; + } + } + return true; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/MakeAbsolute.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/MakeAbsolute.php new file mode 100644 index 000000000..f46ab2630 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/MakeAbsolute.php @@ -0,0 +1,114 @@ +<?php + +// does not support network paths + +class HTMLPurifier_URIFilter_MakeAbsolute extends HTMLPurifier_URIFilter +{ + public $name = 'MakeAbsolute'; + protected $base; + protected $basePathStack = array(); + public function prepare($config) { + $def = $config->getDefinition('URI'); + $this->base = $def->base; + if (is_null($this->base)) { + trigger_error('URI.MakeAbsolute is being ignored due to lack of value for URI.Base configuration', E_USER_WARNING); + return false; + } + $this->base->fragment = null; // fragment is invalid for base URI + $stack = explode('/', $this->base->path); + array_pop($stack); // discard last segment + $stack = $this->_collapseStack($stack); // do pre-parsing + $this->basePathStack = $stack; + return true; + } + public function filter(&$uri, $config, $context) { + if (is_null($this->base)) return true; // abort early + if ( + $uri->path === '' && is_null($uri->scheme) && + is_null($uri->host) && is_null($uri->query) && is_null($uri->fragment) + ) { + // reference to current document + $uri = clone $this->base; + return true; + } + if (!is_null($uri->scheme)) { + // absolute URI already: don't change + if (!is_null($uri->host)) return true; + $scheme_obj = $uri->getSchemeObj($config, $context); + if (!$scheme_obj) { + // scheme not recognized + return false; + } + if (!$scheme_obj->hierarchical) { + // non-hierarchal URI with explicit scheme, don't change + return true; + } + // special case: had a scheme but always is hierarchical and had no authority + } + if (!is_null($uri->host)) { + // network path, don't bother + return true; + } + if ($uri->path === '') { + $uri->path = $this->base->path; + } elseif ($uri->path[0] !== '/') { + // relative path, needs more complicated processing + $stack = explode('/', $uri->path); + $new_stack = array_merge($this->basePathStack, $stack); + if ($new_stack[0] !== '' && !is_null($this->base->host)) { + array_unshift($new_stack, ''); + } + $new_stack = $this->_collapseStack($new_stack); + $uri->path = implode('/', $new_stack); + } else { + // absolute path, but still we should collapse + $uri->path = implode('/', $this->_collapseStack(explode('/', $uri->path))); + } + // re-combine + $uri->scheme = $this->base->scheme; + if (is_null($uri->userinfo)) $uri->userinfo = $this->base->userinfo; + if (is_null($uri->host)) $uri->host = $this->base->host; + if (is_null($uri->port)) $uri->port = $this->base->port; + return true; + } + + /** + * Resolve dots and double-dots in a path stack + */ + private function _collapseStack($stack) { + $result = array(); + $is_folder = false; + for ($i = 0; isset($stack[$i]); $i++) { + $is_folder = false; + // absorb an internally duplicated slash + if ($stack[$i] == '' && $i && isset($stack[$i+1])) continue; + if ($stack[$i] == '..') { + if (!empty($result)) { + $segment = array_pop($result); + if ($segment === '' && empty($result)) { + // error case: attempted to back out too far: + // restore the leading slash + $result[] = ''; + } elseif ($segment === '..') { + $result[] = '..'; // cannot remove .. with .. + } + } else { + // relative path, preserve the double-dots + $result[] = '..'; + } + $is_folder = true; + continue; + } + if ($stack[$i] == '.') { + // silently absorb + $is_folder = true; + continue; + } + $result[] = $stack[$i]; + } + if ($is_folder) $result[] = ''; + return $result; + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/Munge.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/Munge.php new file mode 100644 index 000000000..de695df14 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/Munge.php @@ -0,0 +1,53 @@ +<?php + +class HTMLPurifier_URIFilter_Munge extends HTMLPurifier_URIFilter +{ + public $name = 'Munge'; + public $post = true; + private $target, $parser, $doEmbed, $secretKey; + + protected $replace = array(); + + public function prepare($config) { + $this->target = $config->get('URI.' . $this->name); + $this->parser = new HTMLPurifier_URIParser(); + $this->doEmbed = $config->get('URI.MungeResources'); + $this->secretKey = $config->get('URI.MungeSecretKey'); + return true; + } + public function filter(&$uri, $config, $context) { + if ($context->get('EmbeddedURI', true) && !$this->doEmbed) return true; + + $scheme_obj = $uri->getSchemeObj($config, $context); + if (!$scheme_obj) return true; // ignore unknown schemes, maybe another postfilter did it + if (!$scheme_obj->browsable) return true; // ignore non-browseable schemes, since we can't munge those in a reasonable way + if ($uri->isBenign($config, $context)) return true; // don't redirect if a benign URL + + $this->makeReplace($uri, $config, $context); + $this->replace = array_map('rawurlencode', $this->replace); + + $new_uri = strtr($this->target, $this->replace); + $new_uri = $this->parser->parse($new_uri); + // don't redirect if the target host is the same as the + // starting host + if ($uri->host === $new_uri->host) return true; + $uri = $new_uri; // overwrite + return true; + } + + protected function makeReplace($uri, $config, $context) { + $string = $uri->toString(); + // always available + $this->replace['%s'] = $string; + $this->replace['%r'] = $context->get('EmbeddedURI', true); + $token = $context->get('CurrentToken', true); + $this->replace['%n'] = $token ? $token->name : null; + $this->replace['%m'] = $context->get('CurrentAttr', true); + $this->replace['%p'] = $context->get('CurrentCSSProperty', true); + // not always available + if ($this->secretKey) $this->replace['%t'] = sha1($this->secretKey . ':' . $string); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIFilter/SafeIframe.php b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/SafeIframe.php new file mode 100644 index 000000000..284bb13de --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIFilter/SafeIframe.php @@ -0,0 +1,35 @@ +<?php + +/** + * Implements safety checks for safe iframes. + * + * @warning This filter is *critical* for ensuring that %HTML.SafeIframe + * works safely. + */ +class HTMLPurifier_URIFilter_SafeIframe extends HTMLPurifier_URIFilter +{ + public $name = 'SafeIframe'; + public $always_load = true; + protected $regexp = NULL; + // XXX: The not so good bit about how this is all setup now is we + // can't check HTML.SafeIframe in the 'prepare' step: we have to + // defer till the actual filtering. + public function prepare($config) { + $this->regexp = $config->get('URI.SafeIframeRegexp'); + return true; + } + public function filter(&$uri, $config, $context) { + // check if filter not applicable + if (!$config->get('HTML.SafeIframe')) return true; + // check if the filter should actually trigger + if (!$context->get('EmbeddedURI', true)) return true; + $token = $context->get('CurrentToken', true); + if (!($token && $token->name == 'iframe')) return true; + // check if we actually have some whitelists enabled + if ($this->regexp === null) return false; + // actually check the whitelists + return preg_match($this->regexp, $uri->toString()); + } +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIParser.php b/lib/htmlpurifier/library/HTMLPurifier/URIParser.php new file mode 100644 index 000000000..7179e4ab8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIParser.php @@ -0,0 +1,70 @@ +<?php + +/** + * Parses a URI into the components and fragment identifier as specified + * by RFC 3986. + */ +class HTMLPurifier_URIParser +{ + + /** + * Instance of HTMLPurifier_PercentEncoder to do normalization with. + */ + protected $percentEncoder; + + public function __construct() { + $this->percentEncoder = new HTMLPurifier_PercentEncoder(); + } + + /** + * Parses a URI. + * @param $uri string URI to parse + * @return HTMLPurifier_URI representation of URI. This representation has + * not been validated yet and may not conform to RFC. + */ + public function parse($uri) { + + $uri = $this->percentEncoder->normalize($uri); + + // Regexp is as per Appendix B. + // Note that ["<>] are an addition to the RFC's recommended + // characters, because they represent external delimeters. + $r_URI = '!'. + '(([^:/?#"<>]+):)?'. // 2. Scheme + '(//([^/?#"<>]*))?'. // 4. Authority + '([^?#"<>]*)'. // 5. Path + '(\?([^#"<>]*))?'. // 7. Query + '(#([^"<>]*))?'. // 8. Fragment + '!'; + + $matches = array(); + $result = preg_match($r_URI, $uri, $matches); + + if (!$result) return false; // *really* invalid URI + + // seperate out parts + $scheme = !empty($matches[1]) ? $matches[2] : null; + $authority = !empty($matches[3]) ? $matches[4] : null; + $path = $matches[5]; // always present, can be empty + $query = !empty($matches[6]) ? $matches[7] : null; + $fragment = !empty($matches[8]) ? $matches[9] : null; + + // further parse authority + if ($authority !== null) { + $r_authority = "/^((.+?)@)?(\[[^\]]+\]|[^:]*)(:(\d*))?/"; + $matches = array(); + preg_match($r_authority, $authority, $matches); + $userinfo = !empty($matches[1]) ? $matches[2] : null; + $host = !empty($matches[3]) ? $matches[3] : ''; + $port = !empty($matches[4]) ? (int) $matches[5] : null; + } else { + $port = $host = $userinfo = null; + } + + return new HTMLPurifier_URI( + $scheme, $userinfo, $host, $port, $path, $query, $fragment); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php new file mode 100644 index 000000000..7be958143 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme.php @@ -0,0 +1,95 @@ +<?php + +/** + * Validator for the components of a URI for a specific scheme + */ +abstract class HTMLPurifier_URIScheme +{ + + /** + * Scheme's default port (integer). If an explicit port number is + * specified that coincides with the default port, it will be + * elided. + */ + public $default_port = null; + + /** + * Whether or not URIs of this schem are locatable by a browser + * http and ftp are accessible, while mailto and news are not. + */ + public $browsable = false; + + /** + * Whether or not data transmitted over this scheme is encrypted. + * https is secure, http is not. + */ + public $secure = false; + + /** + * Whether or not the URI always uses <hier_part>, resolves edge cases + * with making relative URIs absolute + */ + public $hierarchical = false; + + /** + * Whether or not the URI may omit a hostname when the scheme is + * explicitly specified, ala file:///path/to/file. As of writing, + * 'file' is the only scheme that browsers support his properly. + */ + public $may_omit_host = false; + + /** + * Validates the components of a URI for a specific scheme. + * @param $uri Reference to a HTMLPurifier_URI object + * @param $config HTMLPurifier_Config object + * @param $context HTMLPurifier_Context object + * @return Bool success or failure + */ + public abstract function doValidate(&$uri, $config, $context); + + /** + * Public interface for validating components of a URI. Performs a + * bunch of default actions. Don't overload this method. + * @param $uri Reference to a HTMLPurifier_URI object + * @param $config HTMLPurifier_Config object + * @param $context HTMLPurifier_Context object + * @return Bool success or failure + */ + public function validate(&$uri, $config, $context) { + if ($this->default_port == $uri->port) $uri->port = null; + // kludge: browsers do funny things when the scheme but not the + // authority is set + if (!$this->may_omit_host && + // if the scheme is present, a missing host is always in error + (!is_null($uri->scheme) && ($uri->host === '' || is_null($uri->host))) || + // if the scheme is not present, a *blank* host is in error, + // since this translates into '///path' which most browsers + // interpret as being 'http://path'. + (is_null($uri->scheme) && $uri->host === '') + ) { + do { + if (is_null($uri->scheme)) { + if (substr($uri->path, 0, 2) != '//') { + $uri->host = null; + break; + } + // URI is '////path', so we cannot nullify the + // host to preserve semantics. Try expanding the + // hostname instead (fall through) + } + // first see if we can manually insert a hostname + $host = $config->get('URI.Host'); + if (!is_null($host)) { + $uri->host = $host; + } else { + // we can't do anything sensible, reject the URL. + return false; + } + } while (false); + } + return $this->doValidate($uri, $config, $context); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/data.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/data.php new file mode 100644 index 000000000..a5c43989e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/data.php @@ -0,0 +1,96 @@ +<?php + +/** + * Implements data: URI for base64 encoded images supported by GD. + */ +class HTMLPurifier_URIScheme_data extends HTMLPurifier_URIScheme { + + public $browsable = true; + public $allowed_types = array( + // you better write validation code for other types if you + // decide to allow them + 'image/jpeg' => true, + 'image/gif' => true, + 'image/png' => true, + ); + // this is actually irrelevant since we only write out the path + // component + public $may_omit_host = true; + + public function doValidate(&$uri, $config, $context) { + $result = explode(',', $uri->path, 2); + $is_base64 = false; + $charset = null; + $content_type = null; + if (count($result) == 2) { + list($metadata, $data) = $result; + // do some legwork on the metadata + $metas = explode(';', $metadata); + while(!empty($metas)) { + $cur = array_shift($metas); + if ($cur == 'base64') { + $is_base64 = true; + break; + } + if (substr($cur, 0, 8) == 'charset=') { + // doesn't match if there are arbitrary spaces, but + // whatever dude + if ($charset !== null) continue; // garbage + $charset = substr($cur, 8); // not used + } else { + if ($content_type !== null) continue; // garbage + $content_type = $cur; + } + } + } else { + $data = $result[0]; + } + if ($content_type !== null && empty($this->allowed_types[$content_type])) { + return false; + } + if ($charset !== null) { + // error; we don't allow plaintext stuff + $charset = null; + } + $data = rawurldecode($data); + if ($is_base64) { + $raw_data = base64_decode($data); + } else { + $raw_data = $data; + } + // XXX probably want to refactor this into a general mechanism + // for filtering arbitrary content types + $file = tempnam("/tmp", ""); + file_put_contents($file, $raw_data); + if (function_exists('exif_imagetype')) { + $image_code = exif_imagetype($file); + } elseif (function_exists('getimagesize')) { + set_error_handler(array($this, 'muteErrorHandler')); + $info = getimagesize($file); + restore_error_handler(); + if ($info == false) return false; + $image_code = $info[2]; + } else { + trigger_error("could not find exif_imagetype or getimagesize functions", E_USER_ERROR); + } + $real_content_type = image_type_to_mime_type($image_code); + if ($real_content_type != $content_type) { + // we're nice guys; if the content type is something else we + // support, change it over + if (empty($this->allowed_types[$real_content_type])) return false; + $content_type = $real_content_type; + } + // ok, it's kosher, rewrite what we need + $uri->userinfo = null; + $uri->host = null; + $uri->port = null; + $uri->fragment = null; + $uri->query = null; + $uri->path = "$content_type;base64," . base64_encode($raw_data); + return true; + } + + public function muteErrorHandler($errno, $errstr) {} + +} + diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/file.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/file.php new file mode 100644 index 000000000..d74a3f198 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/file.php @@ -0,0 +1,32 @@ +<?php + +/** + * Validates file as defined by RFC 1630 and RFC 1738. + */ +class HTMLPurifier_URIScheme_file extends HTMLPurifier_URIScheme { + + // Generally file:// URLs are not accessible from most + // machines, so placing them as an img src is incorrect. + public $browsable = false; + + // Basically the *only* URI scheme for which this is true, since + // accessing files on the local machine is very common. In fact, + // browsers on some operating systems don't understand the + // authority, though I hear it is used on Windows to refer to + // network shares. + public $may_omit_host = true; + + public function doValidate(&$uri, $config, $context) { + // Authentication method is not supported + $uri->userinfo = null; + // file:// makes no provisions for accessing the resource + $uri->port = null; + // While it seems to work on Firefox, the querystring has + // no possible effect and is thus stripped. + $uri->query = null; + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/ftp.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/ftp.php new file mode 100644 index 000000000..0fb2abf64 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/ftp.php @@ -0,0 +1,42 @@ +<?php + +/** + * Validates ftp (File Transfer Protocol) URIs as defined by generic RFC 1738. + */ +class HTMLPurifier_URIScheme_ftp extends HTMLPurifier_URIScheme { + + public $default_port = 21; + public $browsable = true; // usually + public $hierarchical = true; + + public function doValidate(&$uri, $config, $context) { + $uri->query = null; + + // typecode check + $semicolon_pos = strrpos($uri->path, ';'); // reverse + if ($semicolon_pos !== false) { + $type = substr($uri->path, $semicolon_pos + 1); // no semicolon + $uri->path = substr($uri->path, 0, $semicolon_pos); + $type_ret = ''; + if (strpos($type, '=') !== false) { + // figure out whether or not the declaration is correct + list($key, $typecode) = explode('=', $type, 2); + if ($key !== 'type') { + // invalid key, tack it back on encoded + $uri->path .= '%3B' . $type; + } elseif ($typecode === 'a' || $typecode === 'i' || $typecode === 'd') { + $type_ret = ";type=$typecode"; + } + } else { + $uri->path .= '%3B' . $type; + } + $uri->path = str_replace(';', '%3B', $uri->path); + $uri->path .= $type_ret; + } + + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/http.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/http.php new file mode 100644 index 000000000..959b8daff --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/http.php @@ -0,0 +1,19 @@ +<?php + +/** + * Validates http (HyperText Transfer Protocol) as defined by RFC 2616 + */ +class HTMLPurifier_URIScheme_http extends HTMLPurifier_URIScheme { + + public $default_port = 80; + public $browsable = true; + public $hierarchical = true; + + public function doValidate(&$uri, $config, $context) { + $uri->userinfo = null; + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/https.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/https.php new file mode 100644 index 000000000..159c2874e --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/https.php @@ -0,0 +1,13 @@ +<?php + +/** + * Validates https (Secure HTTP) according to http scheme. + */ +class HTMLPurifier_URIScheme_https extends HTMLPurifier_URIScheme_http { + + public $default_port = 443; + public $secure = true; + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/mailto.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/mailto.php new file mode 100644 index 000000000..9db4cb23f --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/mailto.php @@ -0,0 +1,27 @@ +<?php + +// VERY RELAXED! Shouldn't cause problems, not even Firefox checks if the +// email is valid, but be careful! + +/** + * Validates mailto (for E-mail) according to RFC 2368 + * @todo Validate the email address + * @todo Filter allowed query parameters + */ + +class HTMLPurifier_URIScheme_mailto extends HTMLPurifier_URIScheme { + + public $browsable = false; + public $may_omit_host = true; + + public function doValidate(&$uri, $config, $context) { + $uri->userinfo = null; + $uri->host = null; + $uri->port = null; + // we need to validate path against RFC 2368's addr-spec + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/news.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/news.php new file mode 100644 index 000000000..84a6748d8 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/news.php @@ -0,0 +1,22 @@ +<?php + +/** + * Validates news (Usenet) as defined by generic RFC 1738 + */ +class HTMLPurifier_URIScheme_news extends HTMLPurifier_URIScheme { + + public $browsable = false; + public $may_omit_host = true; + + public function doValidate(&$uri, $config, $context) { + $uri->userinfo = null; + $uri->host = null; + $uri->port = null; + $uri->query = null; + // typecode check needed on path + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URIScheme/nntp.php b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/nntp.php new file mode 100644 index 000000000..4ccea0dfc --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URIScheme/nntp.php @@ -0,0 +1,19 @@ +<?php + +/** + * Validates nntp (Network News Transfer Protocol) as defined by generic RFC 1738 + */ +class HTMLPurifier_URIScheme_nntp extends HTMLPurifier_URIScheme { + + public $default_port = 119; + public $browsable = false; + + public function doValidate(&$uri, $config, $context) { + $uri->userinfo = null; + $uri->query = null; + return true; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/URISchemeRegistry.php b/lib/htmlpurifier/library/HTMLPurifier/URISchemeRegistry.php new file mode 100644 index 000000000..576bf7b6d --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/URISchemeRegistry.php @@ -0,0 +1,68 @@ +<?php + +/** + * Registry for retrieving specific URI scheme validator objects. + */ +class HTMLPurifier_URISchemeRegistry +{ + + /** + * Retrieve sole instance of the registry. + * @param $prototype Optional prototype to overload sole instance with, + * or bool true to reset to default registry. + * @note Pass a registry object $prototype with a compatible interface and + * the function will copy it and return it all further times. + */ + public static function instance($prototype = null) { + static $instance = null; + if ($prototype !== null) { + $instance = $prototype; + } elseif ($instance === null || $prototype == true) { + $instance = new HTMLPurifier_URISchemeRegistry(); + } + return $instance; + } + + /** + * Cache of retrieved schemes. + */ + protected $schemes = array(); + + /** + * Retrieves a scheme validator object + * @param $scheme String scheme name like http or mailto + * @param $config HTMLPurifier_Config object + * @param $config HTMLPurifier_Context object + */ + public function getScheme($scheme, $config, $context) { + if (!$config) $config = HTMLPurifier_Config::createDefault(); + + // important, otherwise attacker could include arbitrary file + $allowed_schemes = $config->get('URI.AllowedSchemes'); + if (!$config->get('URI.OverrideAllowedSchemes') && + !isset($allowed_schemes[$scheme]) + ) { + return; + } + + if (isset($this->schemes[$scheme])) return $this->schemes[$scheme]; + if (!isset($allowed_schemes[$scheme])) return; + + $class = 'HTMLPurifier_URIScheme_' . $scheme; + if (!class_exists($class)) return; + $this->schemes[$scheme] = new $class(); + return $this->schemes[$scheme]; + } + + /** + * Registers a custom scheme to the cache, bypassing reflection. + * @param $scheme Scheme name + * @param $scheme_obj HTMLPurifier_URIScheme object + */ + public function register($scheme, $scheme_obj) { + $this->schemes[$scheme] = $scheme_obj; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/UnitConverter.php b/lib/htmlpurifier/library/HTMLPurifier/UnitConverter.php new file mode 100644 index 000000000..545d42622 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/UnitConverter.php @@ -0,0 +1,254 @@ +<?php + +/** + * Class for converting between different unit-lengths as specified by + * CSS. + */ +class HTMLPurifier_UnitConverter +{ + + const ENGLISH = 1; + const METRIC = 2; + const DIGITAL = 3; + + /** + * Units information array. Units are grouped into measuring systems + * (English, Metric), and are assigned an integer representing + * the conversion factor between that unit and the smallest unit in + * the system. Numeric indexes are actually magical constants that + * encode conversion data from one system to the next, with a O(n^2) + * constraint on memory (this is generally not a problem, since + * the number of measuring systems is small.) + */ + protected static $units = array( + self::ENGLISH => array( + 'px' => 3, // This is as per CSS 2.1 and Firefox. Your mileage may vary + 'pt' => 4, + 'pc' => 48, + 'in' => 288, + self::METRIC => array('pt', '0.352777778', 'mm'), + ), + self::METRIC => array( + 'mm' => 1, + 'cm' => 10, + self::ENGLISH => array('mm', '2.83464567', 'pt'), + ), + ); + + /** + * Minimum bcmath precision for output. + */ + protected $outputPrecision; + + /** + * Bcmath precision for internal calculations. + */ + protected $internalPrecision; + + /** + * Whether or not BCMath is available + */ + private $bcmath; + + public function __construct($output_precision = 4, $internal_precision = 10, $force_no_bcmath = false) { + $this->outputPrecision = $output_precision; + $this->internalPrecision = $internal_precision; + $this->bcmath = !$force_no_bcmath && function_exists('bcmul'); + } + + /** + * Converts a length object of one unit into another unit. + * @param HTMLPurifier_Length $length + * Instance of HTMLPurifier_Length to convert. You must validate() + * it before passing it here! + * @param string $to_unit + * Unit to convert to. + * @note + * About precision: This conversion function pays very special + * attention to the incoming precision of values and attempts + * to maintain a number of significant figure. Results are + * fairly accurate up to nine digits. Some caveats: + * - If a number is zero-padded as a result of this significant + * figure tracking, the zeroes will be eliminated. + * - If a number contains less than four sigfigs ($outputPrecision) + * and this causes some decimals to be excluded, those + * decimals will be added on. + */ + public function convert($length, $to_unit) { + + if (!$length->isValid()) return false; + + $n = $length->getN(); + $unit = $length->getUnit(); + + if ($n === '0' || $unit === false) { + return new HTMLPurifier_Length('0', false); + } + + $state = $dest_state = false; + foreach (self::$units as $k => $x) { + if (isset($x[$unit])) $state = $k; + if (isset($x[$to_unit])) $dest_state = $k; + } + if (!$state || !$dest_state) return false; + + // Some calculations about the initial precision of the number; + // this will be useful when we need to do final rounding. + $sigfigs = $this->getSigFigs($n); + if ($sigfigs < $this->outputPrecision) $sigfigs = $this->outputPrecision; + + // BCMath's internal precision deals only with decimals. Use + // our default if the initial number has no decimals, or increase + // it by how ever many decimals, thus, the number of guard digits + // will always be greater than or equal to internalPrecision. + $log = (int) floor(log(abs($n), 10)); + $cp = ($log < 0) ? $this->internalPrecision - $log : $this->internalPrecision; // internal precision + + for ($i = 0; $i < 2; $i++) { + + // Determine what unit IN THIS SYSTEM we need to convert to + if ($dest_state === $state) { + // Simple conversion + $dest_unit = $to_unit; + } else { + // Convert to the smallest unit, pending a system shift + $dest_unit = self::$units[$state][$dest_state][0]; + } + + // Do the conversion if necessary + if ($dest_unit !== $unit) { + $factor = $this->div(self::$units[$state][$unit], self::$units[$state][$dest_unit], $cp); + $n = $this->mul($n, $factor, $cp); + $unit = $dest_unit; + } + + // Output was zero, so bail out early. Shouldn't ever happen. + if ($n === '') { + $n = '0'; + $unit = $to_unit; + break; + } + + // It was a simple conversion, so bail out + if ($dest_state === $state) { + break; + } + + if ($i !== 0) { + // Conversion failed! Apparently, the system we forwarded + // to didn't have this unit. This should never happen! + return false; + } + + // Pre-condition: $i == 0 + + // Perform conversion to next system of units + $n = $this->mul($n, self::$units[$state][$dest_state][1], $cp); + $unit = self::$units[$state][$dest_state][2]; + $state = $dest_state; + + // One more loop around to convert the unit in the new system. + + } + + // Post-condition: $unit == $to_unit + if ($unit !== $to_unit) return false; + + // Useful for debugging: + //echo "<pre>n"; + //echo "$n\nsigfigs = $sigfigs\nnew_log = $new_log\nlog = $log\nrp = $rp\n</pre>\n"; + + $n = $this->round($n, $sigfigs); + if (strpos($n, '.') !== false) $n = rtrim($n, '0'); + $n = rtrim($n, '.'); + + return new HTMLPurifier_Length($n, $unit); + } + + /** + * Returns the number of significant figures in a string number. + * @param string $n Decimal number + * @return int number of sigfigs + */ + public function getSigFigs($n) { + $n = ltrim($n, '0+-'); + $dp = strpos($n, '.'); // decimal position + if ($dp === false) { + $sigfigs = strlen(rtrim($n, '0')); + } else { + $sigfigs = strlen(ltrim($n, '0.')); // eliminate extra decimal character + if ($dp !== 0) $sigfigs--; + } + return $sigfigs; + } + + /** + * Adds two numbers, using arbitrary precision when available. + */ + private function add($s1, $s2, $scale) { + if ($this->bcmath) return bcadd($s1, $s2, $scale); + else return $this->scale($s1 + $s2, $scale); + } + + /** + * Multiples two numbers, using arbitrary precision when available. + */ + private function mul($s1, $s2, $scale) { + if ($this->bcmath) return bcmul($s1, $s2, $scale); + else return $this->scale($s1 * $s2, $scale); + } + + /** + * Divides two numbers, using arbitrary precision when available. + */ + private function div($s1, $s2, $scale) { + if ($this->bcmath) return bcdiv($s1, $s2, $scale); + else return $this->scale($s1 / $s2, $scale); + } + + /** + * Rounds a number according to the number of sigfigs it should have, + * using arbitrary precision when available. + */ + private function round($n, $sigfigs) { + $new_log = (int) floor(log(abs($n), 10)); // Number of digits left of decimal - 1 + $rp = $sigfigs - $new_log - 1; // Number of decimal places needed + $neg = $n < 0 ? '-' : ''; // Negative sign + if ($this->bcmath) { + if ($rp >= 0) { + $n = bcadd($n, $neg . '0.' . str_repeat('0', $rp) . '5', $rp + 1); + $n = bcdiv($n, '1', $rp); + } else { + // This algorithm partially depends on the standardized + // form of numbers that comes out of bcmath. + $n = bcadd($n, $neg . '5' . str_repeat('0', $new_log - $sigfigs), 0); + $n = substr($n, 0, $sigfigs + strlen($neg)) . str_repeat('0', $new_log - $sigfigs + 1); + } + return $n; + } else { + return $this->scale(round($n, $sigfigs - $new_log - 1), $rp + 1); + } + } + + /** + * Scales a float to $scale digits right of decimal point, like BCMath. + */ + private function scale($r, $scale) { + if ($scale < 0) { + // The f sprintf type doesn't support negative numbers, so we + // need to cludge things manually. First get the string. + $r = sprintf('%.0f', (float) $r); + // Due to floating point precision loss, $r will more than likely + // look something like 4652999999999.9234. We grab one more digit + // than we need to precise from $r and then use that to round + // appropriately. + $precise = (string) round(substr($r, 0, strlen($r) + $scale), -1); + // Now we return it, truncating the zero that was rounded off. + return substr($precise, 0, -1) . str_repeat('0', -$scale + 1); + } + return sprintf('%.' . $scale . 'f', (float) $r); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParser.php b/lib/htmlpurifier/library/HTMLPurifier/VarParser.php new file mode 100644 index 000000000..68e72ae86 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParser.php @@ -0,0 +1,154 @@ +<?php + +/** + * Parses string representations into their corresponding native PHP + * variable type. The base implementation does a simple type-check. + */ +class HTMLPurifier_VarParser +{ + + const STRING = 1; + const ISTRING = 2; + const TEXT = 3; + const ITEXT = 4; + const INT = 5; + const FLOAT = 6; + const BOOL = 7; + const LOOKUP = 8; + const ALIST = 9; + const HASH = 10; + const MIXED = 11; + + /** + * Lookup table of allowed types. Mainly for backwards compatibility, but + * also convenient for transforming string type names to the integer constants. + */ + static public $types = array( + 'string' => self::STRING, + 'istring' => self::ISTRING, + 'text' => self::TEXT, + 'itext' => self::ITEXT, + 'int' => self::INT, + 'float' => self::FLOAT, + 'bool' => self::BOOL, + 'lookup' => self::LOOKUP, + 'list' => self::ALIST, + 'hash' => self::HASH, + 'mixed' => self::MIXED + ); + + /** + * Lookup table of types that are string, and can have aliases or + * allowed value lists. + */ + static public $stringTypes = array( + self::STRING => true, + self::ISTRING => true, + self::TEXT => true, + self::ITEXT => true, + ); + + /** + * Validate a variable according to type. Throws + * HTMLPurifier_VarParserException if invalid. + * It may return NULL as a valid type if $allow_null is true. + * + * @param $var Variable to validate + * @param $type Type of variable, see HTMLPurifier_VarParser->types + * @param $allow_null Whether or not to permit null as a value + * @return Validated and type-coerced variable + */ + final public function parse($var, $type, $allow_null = false) { + if (is_string($type)) { + if (!isset(HTMLPurifier_VarParser::$types[$type])) { + throw new HTMLPurifier_VarParserException("Invalid type '$type'"); + } else { + $type = HTMLPurifier_VarParser::$types[$type]; + } + } + $var = $this->parseImplementation($var, $type, $allow_null); + if ($allow_null && $var === null) return null; + // These are basic checks, to make sure nothing horribly wrong + // happened in our implementations. + switch ($type) { + case (self::STRING): + case (self::ISTRING): + case (self::TEXT): + case (self::ITEXT): + if (!is_string($var)) break; + if ($type == self::ISTRING || $type == self::ITEXT) $var = strtolower($var); + return $var; + case (self::INT): + if (!is_int($var)) break; + return $var; + case (self::FLOAT): + if (!is_float($var)) break; + return $var; + case (self::BOOL): + if (!is_bool($var)) break; + return $var; + case (self::LOOKUP): + case (self::ALIST): + case (self::HASH): + if (!is_array($var)) break; + if ($type === self::LOOKUP) { + foreach ($var as $k) if ($k !== true) $this->error('Lookup table contains value other than true'); + } elseif ($type === self::ALIST) { + $keys = array_keys($var); + if (array_keys($keys) !== $keys) $this->error('Indices for list are not uniform'); + } + return $var; + case (self::MIXED): + return $var; + default: + $this->errorInconsistent(get_class($this), $type); + } + $this->errorGeneric($var, $type); + } + + /** + * Actually implements the parsing. Base implementation is to not + * do anything to $var. Subclasses should overload this! + */ + protected function parseImplementation($var, $type, $allow_null) { + return $var; + } + + /** + * Throws an exception. + */ + protected function error($msg) { + throw new HTMLPurifier_VarParserException($msg); + } + + /** + * Throws an inconsistency exception. + * @note This should not ever be called. It would be called if we + * extend the allowed values of HTMLPurifier_VarParser without + * updating subclasses. + */ + protected function errorInconsistent($class, $type) { + throw new HTMLPurifier_Exception("Inconsistency in $class: ".HTMLPurifier_VarParser::getTypeName($type)." not implemented"); + } + + /** + * Generic error for if a type didn't work. + */ + protected function errorGeneric($var, $type) { + $vtype = gettype($var); + $this->error("Expected type ".HTMLPurifier_VarParser::getTypeName($type).", got $vtype"); + } + + static public function getTypeName($type) { + static $lookup; + if (!$lookup) { + // Lazy load the alternative lookup table + $lookup = array_flip(HTMLPurifier_VarParser::$types); + } + if (!isset($lookup[$type])) return 'unknown'; + return $lookup[$type]; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParser/Flexible.php b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Flexible.php new file mode 100644 index 000000000..21b87675a --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Flexible.php @@ -0,0 +1,103 @@ +<?php + +/** + * Performs safe variable parsing based on types which can be used by + * users. This may not be able to represent all possible data inputs, + * however. + */ +class HTMLPurifier_VarParser_Flexible extends HTMLPurifier_VarParser +{ + + protected function parseImplementation($var, $type, $allow_null) { + if ($allow_null && $var === null) return null; + switch ($type) { + // Note: if code "breaks" from the switch, it triggers a generic + // exception to be thrown. Specific errors can be specifically + // done here. + case self::MIXED : + case self::ISTRING : + case self::STRING : + case self::TEXT : + case self::ITEXT : + return $var; + case self::INT : + if (is_string($var) && ctype_digit($var)) $var = (int) $var; + return $var; + case self::FLOAT : + if ((is_string($var) && is_numeric($var)) || is_int($var)) $var = (float) $var; + return $var; + case self::BOOL : + if (is_int($var) && ($var === 0 || $var === 1)) { + $var = (bool) $var; + } elseif (is_string($var)) { + if ($var == 'on' || $var == 'true' || $var == '1') { + $var = true; + } elseif ($var == 'off' || $var == 'false' || $var == '0') { + $var = false; + } else { + throw new HTMLPurifier_VarParserException("Unrecognized value '$var' for $type"); + } + } + return $var; + case self::ALIST : + case self::HASH : + case self::LOOKUP : + if (is_string($var)) { + // special case: technically, this is an array with + // a single empty string item, but having an empty + // array is more intuitive + if ($var == '') return array(); + if (strpos($var, "\n") === false && strpos($var, "\r") === false) { + // simplistic string to array method that only works + // for simple lists of tag names or alphanumeric characters + $var = explode(',',$var); + } else { + $var = preg_split('/(,|[\n\r]+)/', $var); + } + // remove spaces + foreach ($var as $i => $j) $var[$i] = trim($j); + if ($type === self::HASH) { + // key:value,key2:value2 + $nvar = array(); + foreach ($var as $keypair) { + $c = explode(':', $keypair, 2); + if (!isset($c[1])) continue; + $nvar[trim($c[0])] = trim($c[1]); + } + $var = $nvar; + } + } + if (!is_array($var)) break; + $keys = array_keys($var); + if ($keys === array_keys($keys)) { + if ($type == self::ALIST) return $var; + elseif ($type == self::LOOKUP) { + $new = array(); + foreach ($var as $key) { + $new[$key] = true; + } + return $new; + } else break; + } + if ($type === self::ALIST) { + trigger_error("Array list did not have consecutive integer indexes", E_USER_WARNING); + return array_values($var); + } + if ($type === self::LOOKUP) { + foreach ($var as $key => $value) { + if ($value !== true) { + trigger_error("Lookup array has non-true value at key '$key'; maybe your input array was not indexed numerically", E_USER_WARNING); + } + $var[$key] = true; + } + } + return $var; + default: + $this->errorInconsistent(__CLASS__, $type); + } + $this->errorGeneric($var, $type); + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParser/Native.php b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Native.php new file mode 100644 index 000000000..b02a6de54 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParser/Native.php @@ -0,0 +1,26 @@ +<?php + +/** + * This variable parser uses PHP's internal code engine. Because it does + * this, it can represent all inputs; however, it is dangerous and cannot + * be used by users. + */ +class HTMLPurifier_VarParser_Native extends HTMLPurifier_VarParser +{ + + protected function parseImplementation($var, $type, $allow_null) { + return $this->evalExpression($var); + } + + protected function evalExpression($expr) { + $var = null; + $result = eval("\$var = $expr;"); + if ($result === false) { + throw new HTMLPurifier_VarParserException("Fatal error in evaluated code"); + } + return $var; + } + +} + +// vim: et sw=4 sts=4 diff --git a/lib/htmlpurifier/library/HTMLPurifier/VarParserException.php b/lib/htmlpurifier/library/HTMLPurifier/VarParserException.php new file mode 100644 index 000000000..5df341495 --- /dev/null +++ b/lib/htmlpurifier/library/HTMLPurifier/VarParserException.php @@ -0,0 +1,11 @@ +<?php + +/** + * Exception type for HTMLPurifier_VarParser + */ +class HTMLPurifier_VarParserException extends HTMLPurifier_Exception +{ + +} + +// vim: et sw=4 sts=4 |