maintenance/renderDump.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) 2006 Brion Vibber <brion@pobox.com>
  10  * https://www.mediawiki.org/
  11  *
  12  * This program is free software; you can redistribute it and/or modify
  13  * it under the terms of the GNU General Public License as published by
  14  * the Free Software Foundation; either version 2 of the License, or
  15  * (at your option) any later version.
  16  *
  17  * This program is distributed in the hope that it will be useful,
  18  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  19  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  20  * GNU General Public License for more details.
  21  *
  22  * You should have received a copy of the GNU General Public License along
  23  * with this program; if not, write to the Free Software Foundation, Inc.,
  24  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  25  * http://www.gnu.org/copyleft/gpl.html
  26  *
  27  * @file
  28  * @ingroup Maintenance
  29  */
  30
  31 require_once __DIR__ . '/Maintenance.php';
  32
  33 /**
  34  * Maintenance script that takes page text out of an XML dump file
  35  * and render basic HTML out to files.
  36  *
  37  * @ingroup Maintenance
  38  */
  39 class DumpRenderer extends Maintenance {
  40
  41         private $count = 0;
  42         private $outputDirectory, $startTime;
  43
  44         public function __construct() {
  45                 parent::__construct();
  46                 $this->mDescription = "Take page text out of an XML dump file and render basic HTML out to files";
  47                 $this->addOption( 'output-dir', 'The directory to output the HTML files to', true, true );
  48                 $this->addOption( 'prefix', 'Prefix for the rendered files (defaults to wiki)', false, true );
  49                 $this->addOption( 'parser', 'Use an alternative parser class', false, true );
  50         }
  51
  52         public function execute() {
  53                 $this->outputDirectory = $this->getOption( 'output-dir' );
  54                 $this->prefix = $this->getOption( 'prefix', 'wiki' );
  55                 $this->startTime = microtime( true );
  56
  57                 if ( $this->hasOption( 'parser' ) ) {
  58                         global $wgParserConf;
  59                         $wgParserConf['class'] = $this->getOption( 'parser' );
  60                         $this->prefix .= "-{$wgParserConf['class']}";
  61                 }
  62
  63                 $source = new ImportStreamSource( $this->getStdin() );
  64                 $importer = new WikiImporter( $source );
  65
  66                 $importer->setRevisionCallback(
  67                         array( &$this, 'handleRevision' ) );
  68
  69                 $importer->doImport();
  70
  71                 $delta = microtime( true ) - $this->startTime;
  72                 $this->error( "Rendered {$this->count} pages in " . round( $delta, 2 ) . " seconds " );
  73                 if ( $delta > 0 ) {
  74                         $this->error( round( $this->count / $delta, 2 ) . " pages/sec" );
  75                 }
  76                 $this->error( "\n" );
  77         }
  78
  79         /**
  80          * Callback function for each revision, turn into HTML and save
  81          * @param Revision $rev
  82          */
  83         public function handleRevision( $rev ) {
  84                 $title = $rev->getTitle();
  85                 if ( !$title ) {
  86                         $this->error( "Got bogus revision with null title!" );
  87
  88                         return;
  89                 }
  90                 $display = $title->getPrefixedText();
  91
  92                 $this->count++;
  93
  94                 $sanitized = rawurlencode( $display );
  95                 $filename = sprintf( "%s/%s-%07d-%s.html",
  96                         $this->outputDirectory,
  97                         $this->prefix,
  98                         $this->count,
  99                         $sanitized );
 100                 $this->output( sprintf( "%s\n", $filename, $display ) );
 101
 102                 $user = new User();
 103                 $options = ParserOptions::newFromUser( $user );
 104
 105                 $content = $rev->getContent();
 106                 $output = $content->getParserOutput( $title, null, $options );
 107
 108                 file_put_contents( $filename,
 109                         "<!DOCTYPE html>\n" .
 110                         "<html lang=\"en\" dir=\"ltr\">\n" .
 111                         "<head>\n" .
 112                         "<meta charset=\"UTF-8\" />\n" .
 113                         "<title>" . htmlspecialchars( $display ) . "</title>\n" .
 114                         "</head>\n" .
 115                         "<body>\n" .
 116                         $output->getText() .
 117                         "</body>\n" .
 118                         "</html>" );
 119         }
 120 }
 121
 122 $maintClass = "DumpRenderer";
 123 require_once RUN_MAINTENANCE_IF_MAIN;