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  * http://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( dirname( __FILE__ ) . '/Maintenance.php' );
  32
  33 class DumpRenderer extends Maintenance {
  34
  35         private $count = 0;
  36         private $outputDirectory, $startTime;
  37
  38         public function __construct() {
  39                 parent::__construct();
  40                 $this->mDescription = "Take page text out of an XML dump file and render basic HTML out to files";
  41                 $this->addOption( 'output-dir', 'The directory to output the HTML files to', true, true );
  42                 $this->addOption( 'prefix', 'Prefix for the rendered files (defaults to wiki)', false, true );
  43                 $this->addOption( 'parser', 'Use an alternative parser class', false, true );
  44         }
  45
  46         public function execute() {
  47                 $this->outputDirectory = $this->getOption( 'output-dir' );
  48                 $this->prefix = $this->getOption( 'prefix', 'wiki' );
  49                 $this->startTime = wfTime();
  50
  51                 if ( $this->hasOption( 'parser' ) ) {
  52                         global $wgParserConf;
  53                         $wgParserConf['class'] = $this->getOption( 'parser' );
  54                         $this->prefix .= "-{$wgParserConf['class']}";
  55                 }
  56
  57                 $source = new ImportStreamSource( $this->getStdin() );
  58                 $importer = new WikiImporter( $source );
  59
  60                 $importer->setRevisionCallback(
  61                         array( &$this, 'handleRevision' ) );
  62
  63                 $importer->doImport();
  64
  65                 $delta = wfTime() - $this->startTime;
  66                 $this->error( "Rendered {$this->count} pages in " . round($delta, 2) . " seconds " );
  67                 if ($delta > 0)
  68                         $this->error( round($this->count / $delta, 2) . " pages/sec" );
  69                 $this->error( "\n" );
  70         }
  71
  72         /**
  73          * Callback function for each revision, turn into HTML and save
  74          * @param $rev Revision
  75          */
  76         public function handleRevision( $rev ) {
  77                 global $wgParserConf;
  78
  79                 $title = $rev->getTitle();
  80                 if ( !$title ) {
  81                         $this->error( "Got bogus revision with null title!" );
  82                         return;
  83                 }
  84                 $display = $title->getPrefixedText();
  85
  86                 $this->count++;
  87
  88                 $sanitized = rawurlencode( $display );
  89                 $filename = sprintf( "%s/%s-%07d-%s.html",
  90                         $this->outputDirectory,
  91                         $this->prefix,
  92                         $this->count,
  93                         $sanitized );
  94                 $this->output( sprintf( "%s\n", $filename, $display ) );
  95
  96                 $user = new User();
  97                 $parser = new $wgParserConf['class']();
  98                 $options = ParserOptions::newFromUser( $user );
  99
 100                 $output = $parser->parse( $rev->getText(), $title, $options );
 101
 102                 file_put_contents( $filename,
 103                         "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\" " .
 104                         "\"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n" .
 105                         "<html xmlns=\"http://www.w3.org/1999/xhtml\">\n" .
 106                         "<head>\n" .
 107                         "<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n" .
 108                         "<title>" . htmlspecialchars( $display ) . "</title>\n" .
 109                         "</head>\n" .
 110                         "<body>\n" .
 111                         $output->getText() .
 112                         "</body>\n" .
 113                         "</html>" );
 114         }
 115 }
 116
 117 $maintClass = "DumpRenderer";
 118 require_once( RUN_MAINTENANCE_IF_MAIN );