maintenance/compareParsers.php

   1 <?php
   2 /**
   3  * Take page text out of an XML dump file and render basic HTML out to files.
   4  * This is *NOT* suitable for publishing or offline use; it's intended for
   5  * running comparative tests of parsing behavior using real-world data.
   6  *
   7  * Templates etc are pulled from the local wiki database, not from the dump.
   8  *
   9  * Copyright (C) 2011 Platonides - http://www.mediawiki.org/
  10  *
  11  * This program is free software; you can redistribute it and/or modify
  12  * it under the terms of the GNU General Public License as published by
  13  * the Free Software Foundation; either version 2 of the License, or
  14  * (at your option) any later version.
  15  *
  16  * This program is distributed in the hope that it will be useful,
  17  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  18  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  19  * GNU General Public License for more details.
  20  *
  21  * You should have received a copy of the GNU General Public License along
  22  * with this program; if not, write to the Free Software Foundation, Inc.,
  23  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  24  * http://www.gnu.org/copyleft/gpl.html
  25  *
  26  * @file
  27  * @ingroup Maintenance
  28  */
  29
  30 require_once( dirname( __FILE__ ) . '/dumpIterator.php' );
  31
  32 class CompareParsers extends DumpIterator {
  33
  34         private $count = 0;
  35         private $startTime;
  36
  37         public function __construct() {
  38                 parent::__construct();
  39                 $this->saveFailed = false;
  40                 $this->mDescription = "Run a file or dump with several parsers";
  41                 $this->addOption( 'parser1', 'The first parser to compare.', true, true );
  42                 $this->addOption( 'parser2', 'The second parser to compare.', true, true );
  43                 $this->addOption( 'tidy', 'Run tidy on the articles.', false, false );
  44                 $this->addOption( 'save-failed', 'Folder in which articles which differ will be stored.', false, true );
  45                 $this->addOption( 'show-diff', 'Show a diff of the two renderings.', false, false );
  46                 $this->addOption( 'diff-bin', 'Binary to use for diffing (can also be provided by DIFF env var).', false, false );
  47                 $this->addOption( 'strip-parameters', 'Remove parameters of html tags to increase readability.', false, false );
  48                 $this->addOption( 'show-parsed-output', 'Show the parsed html if both Parsers give the same output.', false, false );
  49         }
  50
  51         public function checkOptions() {
  52                 if ( $this->hasOption('save-failed') ) {
  53                         $this->saveFailed = $this->getOption('save-failed');
  54                 }
  55
  56                 $this->stripParametersEnabled = $this->hasOption( 'strip-parameters' );
  57                 $this->showParsedOutput = $this->hasOption( 'show-parsed-output' );
  58
  59                 $this->showDiff = $this->hasOption( 'show-diff' );
  60                 if ( $this->showDiff ) {
  61                         $bin = $this->getOption( 'diff-bin', getenv( 'DIFF' ) );
  62                         if ( $bin != '' ) {
  63                                 global $wgDiff;
  64                                 $wgDiff = $bin;
  65                         }
  66                 }
  67
  68                 $user = new User();
  69                 $this->options = ParserOptions::newFromUser( $user );
  70
  71                 if ( $this->hasOption( 'tidy' ) ) {
  72                         global $wgUseTidy;
  73                         if ( !$wgUseTidy ) {
  74                                 $this->error( 'Tidy was requested but $wgUseTidy is not set in LocalSettings.php', true );
  75                         }
  76                         $this->options->setTidy( true );
  77                 }
  78
  79                 $this->failed = 0;
  80         }
  81
  82         public function conclusions() {
  83                 $this->error( "{$this->failed} failed revisions out of {$this->count}" );
  84                 if ($this->count > 0)
  85                         $this->output( " (" . ( $this->failed / $this->count ) . "%)\n" );
  86         }
  87
  88         function stripParameters( $text ) {
  89                 if ( !$this->stripParametersEnabled ) {
  90                         return $text;
  91                 }
  92                 return preg_replace( '/(<a) [^>]+>/', '$1>', $text );
  93         }
  94
  95         /**
  96          * Callback function for each revision, parse with both parsers and compare
  97          * @param $rev Revision
  98          */
  99         public function processRevision( $rev ) {
 100                 $title = $rev->getTitle();
 101
 102                 $parser1Name = $this->getOption( 'parser1' );
 103                 $parser2Name = $this->getOption( 'parser2' );
 104
 105                 self::checkParserLocally( $parser1Name );
 106                 self::checkParserLocally( $parser2Name );
 107
 108                 $parser1 = new $parser1Name();
 109                 $parser2 = new $parser2Name();
 110
 111                 $output1 = $parser1->parse( $rev->getText(), $title, $this->options );
 112                 $output2 = $parser2->parse( $rev->getText(), $title, $this->options );
 113
 114                 if ( $output1->getText() != $output2->getText() ) {
 115                         $this->failed++;
 116                         $this->error( "Parsing for {$title->getPrefixedText()} differs\n" );
 117
 118                         if ( $this->saveFailed ) {
 119                                 file_put_contents( $this->saveFailed . '/' . rawurlencode( $title->getPrefixedText() ) . ".txt", $rev->getText());
 120                         }
 121                         if ( $this->showDiff ) {
 122                                 $this->output( wfDiff( $this->stripParameters( $output1->getText() ), $this->stripParameters( $output2->getText() ), '' ) );
 123                         }
 124                 } else {
 125                         $this->output( $title->getPrefixedText() . "\tOK\n" );
 126                         if ( $this->showParsedOutput ) {
 127                                 $this->output( $this->stripParameters( $output1->getText() ) );
 128                         }
 129                 }
 130         }
 131
 132         private static function checkParserLocally( $parserName ) {
 133                 /* Look for the parser in a file appropiately named in the current folder */
 134                 if ( !class_exists( $parserName ) && file_exists( "$parserName.php" ) ) {
 135                         global $wgAutoloadClasses;
 136                         $wgAutoloadClasses[ $parserName ] = realpath( '.' ) . "/$parserName.php";
 137                 }
 138         }
 139
 140 }
 141
 142 $maintClass = "CompareParsers";
 143 require( RUN_MAINTENANCE_IF_MAIN );