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          */
  45         public static function tidy( $text ) {
  46                 $driver = self::singleton();
  47                 if ( !$driver ) {
  48                         throw new MWException( __METHOD__ .
  49                                 ': tidy is disabled, caller should have checked MWTidy::isEnabled()' );
  50                 }
  51                 return $driver->tidy( $text );
  52         }
  53
  54         /**
  55          * Get CSS modules needed if HTML from the current driver is to be displayed.
  56          *
  57          * This is just a migration tool to allow some changes expected as part of
  58          * Tidy replacement (T89331) to be exposed on the client side via user
  59          * scripts, without actually replacing tidy. See T49673.
  60          *
  61          * @return array
  62          */
  63         public static function getModuleStyles() {
  64                 $driver = self::singleton();
  65                 if ( $driver && $driver instanceof MediaWiki\Tidy\RaggettBase ) {
  66                         return [ 'mediawiki.raggett' ];
  67                 } else {
  68                         return [];
  69                 }
  70         }
  71
  72         /**
  73          * Check HTML for errors, used if $wgValidateAllHtml = true.
  74          *
  75          * @param string $text
  76          * @param string &$errorStr Return the error string
  77          * @return bool Whether the HTML is valid
  78          */
  79         public static function checkErrors( $text, &$errorStr = null ) {
  80                 $driver = self::singleton();
  81                 if ( !$driver ) {
  82                         throw new MWException( __METHOD__ .
  83                                 ': tidy is disabled, caller should have checked MWTidy::isEnabled()' );
  84                 }
  85                 if ( $driver->supportsValidate() ) {
  86                         return $driver->validate( $text, $errorStr );
  87                 } else {
  88                         throw new MWException( __METHOD__ . ": error text return from HHVM tidy is not supported" );
  89                 }
  90         }
  91
  92         public static function isEnabled() {
  93                 return self::singleton() !== false;
  94         }
  95
  96         protected static function singleton() {
  97                 global $wgUseTidy, $wgTidyInternal, $wgTidyConf, $wgDebugTidy, $wgTidyConfig,
  98                         $wgTidyBin, $wgTidyOpts;
  99
 100                 if ( self::$instance === null ) {
 101                         if ( $wgTidyConfig !== null ) {
 102                                 $config = $wgTidyConfig;
 103                         } elseif ( $wgUseTidy ) {
 104                                 // b/c configuration
 105                                 $config = [
 106                                         'tidyConfigFile' => $wgTidyConf,
 107                                         'debugComment' => $wgDebugTidy,
 108                                         'tidyBin' => $wgTidyBin,
 109                                         'tidyCommandLine' => $wgTidyOpts ];
 110                                 if ( $wgTidyInternal ) {
 111                                         if ( wfIsHHVM() ) {
 112                                                 $config['driver'] = 'RaggettInternalHHVM';
 113                                         } else {
 114                                                 $config['driver'] = 'RaggettInternalPHP';
 115                                         }
 116                                 } else {
 117                                         $config['driver'] = 'RaggettExternal';
 118                                 }
 119                         } else {
 120                                 return false;
 121                         }
 122                         switch ( $config['driver'] ) {
 123                                 case 'RaggettInternalHHVM':
 124                                         self::$instance = new MediaWiki\Tidy\RaggettInternalHHVM( $config );
 125                                         break;
 126                                 case 'RaggettInternalPHP':
 127                                         self::$instance = new MediaWiki\Tidy\RaggettInternalPHP( $config );
 128                                         break;
 129                                 case 'RaggettExternal':
 130                                         self::$instance = new MediaWiki\Tidy\RaggettExternal( $config );
 131                                         break;
 132                                 case 'Html5Depurate':
 133                                         self::$instance = new MediaWiki\Tidy\Html5Depurate( $config );
 134                                         break;
 135                                 default:
 136                                         throw new MWException( "Invalid tidy driver: \"{$config['driver']}\"" );
 137                         }
 138                 }
 139                 return self::$instance;
 140         }
 141
 142         /**
 143          * Set the driver to be used. This is for testing.
 144          * @param TidyDriverBase|false|null $instance
 145          */
 146         public static function setInstance( $instance ) {
 147                 self::$instance = $instance;
 148         }
 149
 150         /**
 151          * Destroy the current singleton instance
 152          */
 153         public static function destroySingleton() {
 154                 self::$instance = null;
 155         }
 156 }