includes/parser/MWTidy.php

   1 <?php
   2 /**
   3  * HTML validation and correction
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  * @ingroup Parser
  22  */
  23
  24 /**
  25  * Class to interact with HTML tidy
  26  *
  27  * Either the external tidy program or the in-process tidy extension
  28  * will be used depending on availability. Override the default
  29  * $wgTidyInternal setting to disable the internal if it's not working.
  30  *
  31  * @ingroup Parser
  32  */
  33 class MWTidy {
  34         private static $instance;
  35
  36         /**
  37          * Interface with html tidy.
  38          * If tidy isn't able to correct the markup, the original will be
  39          * returned in all its glory with a warning comment appended.
  40          *
  41          * @param string $text HTML input fragment. This should not contain a
  42          *                     <body> or <html> tag.
  43          * @return string Corrected HTML output
  44          * @throws MWException
  45          */
  46         public static function tidy( $text ) {
  47                 $driver = self::singleton();
  48                 if ( !$driver ) {
  49                         throw new MWException( __METHOD__ .
  50                                 ': tidy is disabled, caller should have checked MWTidy::isEnabled()' );
  51                 }
  52                 return $driver->tidy( $text );
  53         }
  54
  55         /**
  56          * Check HTML for errors, used if $wgValidateAllHtml = true.
  57          *
  58          * @param string $text
  59          * @param string &$errorStr Return the error string
  60          * @return bool Whether the HTML is valid
  61          * @throws MWException
  62          */
  63         public static function checkErrors( $text, &$errorStr = null ) {
  64                 $driver = self::singleton();
  65                 if ( !$driver ) {
  66                         throw new MWException( __METHOD__ .
  67                                 ': tidy is disabled, caller should have checked MWTidy::isEnabled()' );
  68                 }
  69                 if ( $driver->supportsValidate() ) {
  70                         return $driver->validate( $text, $errorStr );
  71                 } else {
  72                         throw new MWException( __METHOD__ . ": error text return from HHVM tidy is not supported" );
  73                 }
  74         }
  75
  76         /**
  77          * @return bool
  78          */
  79         public static function isEnabled() {
  80                 return self::singleton() !== false;
  81         }
  82
  83         /**
  84          * @return bool|\MediaWiki\Tidy\TidyDriverBase
  85          */
  86         protected static function singleton() {
  87                 global $wgUseTidy, $wgTidyInternal, $wgTidyConf, $wgDebugTidy, $wgTidyConfig,
  88                         $wgTidyBin, $wgTidyOpts;
  89
  90                 if ( self::$instance === null ) {
  91                         if ( $wgTidyConfig !== null ) {
  92                                 $config = $wgTidyConfig;
  93                         } elseif ( $wgUseTidy ) {
  94                                 // b/c configuration
  95                                 $config = [
  96                                         'tidyConfigFile' => $wgTidyConf,
  97                                         'debugComment' => $wgDebugTidy,
  98                                         'tidyBin' => $wgTidyBin,
  99                                         'tidyCommandLine' => $wgTidyOpts ];
 100                                 if ( $wgTidyInternal ) {
 101                                         if ( wfIsHHVM() ) {
 102                                                 $config['driver'] = 'RaggettInternalHHVM';
 103                                         } else {
 104                                                 $config['driver'] = 'RaggettInternalPHP';
 105                                         }
 106                                 } else {
 107                                         $config['driver'] = 'RaggettExternal';
 108                                 }
 109                         } else {
 110                                 return false;
 111                         }
 112                         self::$instance = self::factory( $config );
 113                 }
 114                 return self::$instance;
 115         }
 116
 117         /**
 118          * Create a new Tidy driver object from configuration.
 119          * @see $wgTidyConfig
 120          * @param array $config
 121          * @return bool|\MediaWiki\Tidy\TidyDriverBase
 122          * @throws MWException
 123          */
 124         public static function factory( array $config ) {
 125                 switch ( $config['driver'] ) {
 126                         case 'RaggettInternalHHVM':
 127                                 $instance = new MediaWiki\Tidy\RaggettInternalHHVM( $config );
 128                                 break;
 129                         case 'RaggettInternalPHP':
 130                                 $instance = new MediaWiki\Tidy\RaggettInternalPHP( $config );
 131                                 break;
 132                         case 'RaggettExternal':
 133                                 $instance = new MediaWiki\Tidy\RaggettExternal( $config );
 134                                 break;
 135                         case 'Html5Depurate':
 136                                 $instance = new MediaWiki\Tidy\Html5Depurate( $config );
 137                                 break;
 138                         case 'Html5Internal':
 139                                 $instance = new MediaWiki\Tidy\Html5Internal( $config );
 140                                 break;
 141                         case 'disabled':
 142                                 return false;
 143                         default:
 144                                 throw new MWException( "Invalid tidy driver: \"{$config['driver']}\"" );
 145                 }
 146                 return $instance;
 147         }
 148
 149         /**
 150          * Set the driver to be used. This is for testing.
 151          * @param MediaWiki\Tidy\TidyDriverBase|false|null $instance
 152          */
 153         public static function setInstance( $instance ) {
 154                 self::$instance = $instance;
 155         }
 156
 157         /**
 158          * Destroy the current singleton instance
 159          */
 160         public static function destroySingleton() {
 161                 self::$instance = null;
 162         }
 163 }