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 comparitive 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  * @addtogroup Maintenance
  28  */
  29
  30 $optionsWithArgs = array( 'report' );
  31
  32 require_once( 'commandLine.inc' );
  33 require_once( 'SpecialImport.php' );
  34
  35 class DumpRenderer {
  36         function __construct( $dir ) {
  37                 $this->stderr = fopen( "php://stderr", "wt" );
  38                 $this->outputDirectory = $dir;
  39                 $this->count = 0;
  40         }
  41
  42         function handleRevision( $rev ) {
  43                 $title = $rev->getTitle();
  44                 if (!$title) {
  45                         fprintf( $this->stderr, "Got bogus revision with null title!" );
  46                         return;
  47                 }
  48                 $display = $title->getPrefixedText();
  49
  50                 $this->count++;
  51
  52                 $sanitized = rawurlencode( $display );
  53                 $filename = sprintf( "%s/wiki-%07d-%s.html",
  54                         $this->outputDirectory,
  55                         $this->count,
  56                         $sanitized );
  57                 fprintf( $this->stderr, "%s\n", $filename, $display );
  58
  59                 // fixme
  60                 $user = new User();
  61                 $parser = new Parser();
  62                 $options = ParserOptions::newFromUser( $user );
  63
  64                 $output = $parser->parse( $rev->getText(), $title, $options );
  65
  66                 file_put_contents( $filename,
  67                         "<!DOCTYPE html PUBLIC \"-//W3C//DTD XHTML 1.0 Transitional//EN\" " .
  68                         "\"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd\">\n" .
  69                         "<html xmlns=\"http://www.w3.org/1999/xhtml\">\n" .
  70                         "<head>\n" .
  71                         "<meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n" .
  72                         "<title>" . htmlspecialchars( $display ) . "</title>\n" .
  73                         "</head>\n" .
  74                         "<body>\n" .
  75                         $output->getText() .
  76                         "</body>\n" .
  77                         "</html>" );
  78         }
  79
  80         function run() {
  81                 $this->startTime = wfTime();
  82
  83                 $file = fopen( 'php://stdin', 'rt' );
  84                 $source = new ImportStreamSource( $file );
  85                 $importer = new WikiImporter( $source );
  86
  87                 $importer->setRevisionCallback(
  88                         array( &$this, 'handleRevision' ) );
  89
  90                 return $importer->doImport();
  91         }
  92 }
  93
  94 if( isset( $options['output-dir'] ) ) {
  95         $dir = $options['output-dir'];
  96 } else {
  97         wfDie( "Must use --output-dir=/some/dir\n" );
  98 }
  99 $render = new DumpRenderer( $dir );
 100 $render->run();
 101
 102 ?>