library/HTMLPurifier.php

   1 <?php
   2
   3 /*! @mainpage
   4  *
   5  * HTML Purifier is an HTML filter that will take an arbitrary snippet of
   6  * HTML and rigorously test, validate and filter it into a version that
   7  * is safe for output onto webpages. It achieves this by:
   8  *
   9  *  -# Lexing (parsing into tokens) the document,
  10  *  -# Executing various strategies on the tokens:
  11  *      -# Removing all elements not in the whitelist,
  12  *      -# Making the tokens well-formed,
  13  *      -# Fixing the nesting of the nodes, and
  14  *      -# Validating attributes of the nodes; and
  15  *  -# Generating HTML from the purified tokens.
  16  *
  17  * However, most users will only need to interface with the HTMLPurifier
  18  * and HTMLPurifier_Config.
  19  */
  20
  21 /*
  22     HTML Purifier 3.1.0-dev - Standards Compliant HTML Filtering
  23     Copyright (C) 2006-2008 Edward Z. Yang
  24
  25     This library is free software; you can redistribute it and/or
  26     modify it under the terms of the GNU Lesser General Public
  27     License as published by the Free Software Foundation; either
  28     version 2.1 of the License, or (at your option) any later version.
  29
  30     This library is distributed in the hope that it will be useful,
  31     but WITHOUT ANY WARRANTY; without even the implied warranty of
  32     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  33     Lesser General Public License for more details.
  34
  35     You should have received a copy of the GNU Lesser General Public
  36     License along with this library; if not, write to the Free Software
  37     Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
  38  */
  39
  40 /**
  41  * Facade that coordinates HTML Purifier's subsystems in order to purify HTML.
  42  *
  43  * @note There are several points in which configuration can be specified
  44  *       for HTML Purifier.  The precedence of these (from lowest to
  45  *       highest) is as follows:
  46  *          -# Instance: new HTMLPurifier($config)
  47  *          -# Invocation: purify($html, $config)
  48  *       These configurations are entirely independent of each other and
  49  *       are *not* merged (this behavior may change in the future).
  50  *
  51  * @todo We need an easier way to inject strategies using the configuration
  52  *       object.
  53  */
  54 class HTMLPurifier
  55 {
  56
  57     /** Version of HTML Purifier */
  58     public $version = '3.1.0-dev';
  59
  60     /** Constant with version of HTML Purifier */
  61     const VERSION = '3.1.0-dev';
  62
  63     /** Global configuration object */
  64     public $config;
  65
  66     /** Array of extra HTMLPurifier_Filter objects to run on HTML, for backwards compatibility */
  67     private $filters = array();
  68
  69     /** Single instance of HTML Purifier */
  70     private static $instance;
  71
  72     protected $strategy, $generator;
  73
  74     /**
  75      * Resultant HTMLPurifier_Context of last run purification. Is an array
  76      * of contexts if the last called method was purifyArray().
  77      */
  78     public $context;
  79
  80     /**
  81      * Initializes the purifier.
  82      * @param $config Optional HTMLPurifier_Config object for all instances of
  83      *                the purifier, if omitted, a default configuration is
  84      *                supplied (which can be overridden on a per-use basis).
  85      *                The parameter can also be any type that
  86      *                HTMLPurifier_Config::create() supports.
  87      */
  88     public function __construct($config = null) {
  89
  90         $this->config = HTMLPurifier_Config::create($config);
  91
  92         $this->strategy     = new HTMLPurifier_Strategy_Core();
  93         $this->generator    = new HTMLPurifier_Generator();
  94
  95     }
  96
  97     /**
  98      * Adds a filter to process the output. First come first serve
  99      * @param $filter HTMLPurifier_Filter object
 100      */
 101     public function addFilter($filter) {
 102         trigger_error('HTMLPurifier->addFilter() is deprecated, use configuration directives in the Filter namespace or Filter.Custom', E_USER_WARNING);
 103         $this->filters[] = $filter;
 104     }
 105
 106     /**
 107      * Filters an HTML snippet/document to be XSS-free and standards-compliant.
 108      *
 109      * @param $html String of HTML to purify
 110      * @param $config HTMLPurifier_Config object for this operation, if omitted,
 111      *                defaults to the config object specified during this
 112      *                object's construction. The parameter can also be any type
 113      *                that HTMLPurifier_Config::create() supports.
 114      * @return Purified HTML
 115      */
 116     public function purify($html, $config = null) {
 117
 118         // :TODO: make the config merge in, instead of replace
 119         $config = $config ? HTMLPurifier_Config::create($config) : $this->config;
 120
 121         // implementation is partially environment dependant, partially
 122         // configuration dependant
 123         $lexer = HTMLPurifier_Lexer::create($config);
 124
 125         $context = new HTMLPurifier_Context();
 126
 127         // our friendly neighborhood generator, all primed with configuration too!
 128         $this->generator->generateFromTokens(array(), $config, $context);
 129         $context->register('Generator', $this->generator);
 130
 131         // set up global context variables
 132         if ($config->get('Core', 'CollectErrors')) {
 133             // may get moved out if other facilities use it
 134             $language_factory = HTMLPurifier_LanguageFactory::instance();
 135             $language = $language_factory->create($config, $context);
 136             $context->register('Locale', $language);
 137
 138             $error_collector = new HTMLPurifier_ErrorCollector($context);
 139             $context->register('ErrorCollector', $error_collector);
 140         }
 141
 142         // setup id_accumulator context, necessary due to the fact that
 143         // AttrValidator can be called from many places
 144         $id_accumulator = HTMLPurifier_IDAccumulator::build($config, $context);
 145         $context->register('IDAccumulator', $id_accumulator);
 146
 147         $html = HTMLPurifier_Encoder::convertToUTF8($html, $config, $context);
 148
 149         // setup filters
 150         $filter_flags = $config->getBatch('Filter');
 151         $custom_filters = $filter_flags['Custom'];
 152         unset($filter_flags['Custom']);
 153         $filters = array();
 154         foreach ($filter_flags as $filter => $flag) {
 155             if (!$flag) continue;
 156             $class = "HTMLPurifier_Filter_$filter";
 157             $filters[] = new $class;
 158         }
 159         foreach ($custom_filters as $filter) {
 160             // maybe "HTMLPurifier_Filter_$filter", but be consistent with AutoFormat
 161             $filters[] = $filter;
 162         }
 163         $filters = array_merge($filters, $this->filters);
 164         // maybe prepare(), but later
 165
 166         for ($i = 0, $filter_size = count($filters); $i < $filter_size; $i++) {
 167             $html = $filters[$i]->preFilter($html, $config, $context);
 168         }
 169
 170         // purified HTML
 171         $html =
 172             $this->generator->generateFromTokens(
 173                 // list of tokens
 174                 $this->strategy->execute(
 175                     // list of un-purified tokens
 176                     $lexer->tokenizeHTML(
 177                         // un-purified HTML
 178                         $html, $config, $context
 179                     ),
 180                     $config, $context
 181                 ),
 182                 $config, $context
 183             );
 184
 185         for ($i = $filter_size - 1; $i >= 0; $i--) {
 186             $html = $filters[$i]->postFilter($html, $config, $context);
 187         }
 188
 189         $html = HTMLPurifier_Encoder::convertFromUTF8($html, $config, $context);
 190         $this->context =& $context;
 191         return $html;
 192     }
 193
 194     /**
 195      * Filters an array of HTML snippets
 196      * @param $config Optional HTMLPurifier_Config object for this operation.
 197      *                See HTMLPurifier::purify() for more details.
 198      * @return Array of purified HTML
 199      */
 200     public function purifyArray($array_of_html, $config = null) {
 201         $context_array = array();
 202         foreach ($array_of_html as $key => $html) {
 203             $array_of_html[$key] = $this->purify($html, $config);
 204             $context_array[$key] = $this->context;
 205         }
 206         $this->context = $context_array;
 207         return $array_of_html;
 208     }
 209
 210     /**
 211      * Singleton for enforcing just one HTML Purifier in your system
 212      * @param $prototype Optional prototype HTMLPurifier instance to
 213      *                   overload singleton with, or HTMLPurifier_Config
 214      *                   instance to configure the generated version with.
 215      */
 216     public static function instance($prototype = null) {
 217         if (!self::$instance || $prototype) {
 218             if ($prototype instanceof HTMLPurifier) {
 219                 self::$instance = $prototype;
 220             } elseif ($prototype) {
 221                 self::$instance = new HTMLPurifier($prototype);
 222             } else {
 223                 self::$instance = new HTMLPurifier();
 224             }
 225         }
 226         return self::$instance;
 227     }
 228
 229     /**
 230      * @note Backwards compatibility, see instance()
 231      */
 232     public static function getInstance($prototype = null) {
 233         return HTMLPurifier::instance($prototype);
 234     }
 235
 236 }