maintenance/importDump.php

   1 <?php
   2 /**
   3  * Copyright (C) 2005 Brion Vibber <brion@pobox.com>
   4  * http://www.mediawiki.org/
   5  *
   6  * This program is free software; you can redistribute it and/or modify
   7  * it under the terms of the GNU General Public License as published by
   8  * the Free Software Foundation; either version 2 of the License, or
   9  * (at your option) any later version.
  10  *
  11  * This program is distributed in the hope that it will be useful,
  12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  14  * GNU General Public License for more details.
  15  *
  16  * You should have received a copy of the GNU General Public License along
  17  * with this program; if not, write to the Free Software Foundation, Inc.,
  18  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  19  * http://www.gnu.org/copyleft/gpl.html
  20  *
  21  * @file
  22  * @ingroup Maintenance
  23  */
  24
  25 $optionsWithArgs = array( 'report' );
  26
  27 require_once( dirname(__FILE__) . '/commandLine.inc' );
  28 require_once( '7zip.inc' );
  29
  30 /**
  31  * @ingroup Maintenance
  32  */
  33 class BackupReader {
  34         var $reportingInterval = 100;
  35         var $reporting = true;
  36         var $pageCount = 0;
  37         var $revCount  = 0;
  38         var $dryRun    = false;
  39         var $debug     = false;
  40         var $uploads   = false;
  41
  42         function BackupReader() {
  43                 $this->stderr = fopen( "php://stderr", "wt" );
  44         }
  45
  46         function reportPage( $page ) {
  47                 $this->pageCount++;
  48         }
  49
  50         function handleRevision( $rev ) {
  51                 $title = $rev->getTitle();
  52                 if( !$title ) {
  53                         $this->progress( "Got bogus revision with null title!" );
  54                         return;
  55                 }
  56
  57                 $this->revCount++;
  58                 $this->report();
  59
  60                 if( !$this->dryRun ) {
  61                         call_user_func( $this->importCallback, $rev );
  62                 }
  63         }
  64
  65         function handleUpload( $revision ) {
  66                 if( $this->uploads ) {
  67                         $this->uploadCount++;
  68                         //$this->report();
  69                         $this->progress( "upload: " . $revision->getFilename() );
  70
  71                         if( !$this->dryRun ) {
  72                                 // bluuuh hack
  73                                 //call_user_func( $this->uploadCallback, $revision );
  74                                 $dbw = wfGetDB( DB_MASTER );
  75                                 return $dbw->deadlockLoop( array( $revision, 'importUpload' ) );
  76                         }
  77                 }
  78         }
  79
  80         function handleLogItem( $rev ) {
  81                 $this->revCount++;
  82                 $this->report();
  83
  84                 if( !$this->dryRun ) {
  85                         call_user_func( $this->logItemCallback, $rev );
  86                 }
  87         }
  88
  89         function report( $final = false ) {
  90                 if( $final xor ( $this->pageCount % $this->reportingInterval == 0 ) ) {
  91                         $this->showReport();
  92                 }
  93         }
  94
  95         function showReport() {
  96                 if( $this->reporting ) {
  97                         $delta = wfTime() - $this->startTime;
  98                         if( $delta ) {
  99                                 $rate = sprintf("%.2f", $this->pageCount / $delta);
 100                                 $revrate = sprintf("%.2f", $this->revCount / $delta);
 101                         } else {
 102                                 $rate = '-';
 103                                 $revrate = '-';
 104                         }
 105                         # Logs dumps don't have page tallies
 106                         if( $this->pageCount )
 107                                 $this->progress( "$this->pageCount ($rate pages/sec $revrate revs/sec)" );
 108                         else
 109                                 $this->progress( "$this->revCount ($revrate revs/sec)" );
 110                 }
 111                 wfWaitForSlaves(5);
 112         }
 113
 114         function progress( $string ) {
 115                 fwrite( $this->stderr, $string . "\n" );
 116         }
 117
 118         function importFromFile( $filename ) {
 119                 $t = true;
 120                 if( preg_match( '/\.gz$/', $filename ) ) {
 121                         $filename = 'compress.zlib://' . $filename;
 122                 }
 123                 elseif( preg_match( '/\.bz2$/', $filename ) ) {
 124                         $filename = 'compress.bzip2://' . $filename;
 125                 }
 126                 elseif( preg_match( '/\.7z$/', $filename ) ) {
 127                         $filename = 'mediawiki.compress.7z://' . $filename;
 128                         $t = false;
 129                 }
 130
 131                 $file = fopen( $filename, $t ? 'rt' : 't' ); //our 7zip wrapper uses popen, which seems not to like two-letter modes
 132                 return $this->importFromHandle( $file );
 133         }
 134
 135         function importFromStdin() {
 136                 $file = fopen( 'php://stdin', 'rt' );
 137                 return $this->importFromHandle( $file );
 138         }
 139
 140         function importFromHandle( $handle ) {
 141                 $this->startTime = wfTime();
 142
 143                 $source = new ImportStreamSource( $handle );
 144                 $importer = new WikiImporter( $source );
 145
 146                 $importer->setDebug( $this->debug );
 147                 $importer->setPageCallback( array( &$this, 'reportPage' ) );
 148                 $this->importCallback =  $importer->setRevisionCallback(
 149                         array( &$this, 'handleRevision' ) );
 150                 $this->uploadCallback = $importer->setUploadCallback(
 151                         array( &$this, 'handleUpload' ) );
 152                 $this->logItemCallback = $importer->setLogItemCallback(
 153                         array( &$this, 'handleLogItem' ) );
 154
 155                 return $importer->doImport();
 156         }
 157 }
 158
 159 if( wfReadOnly() ) {
 160         wfDie( "Wiki is in read-only mode; you'll need to disable it for import to work.\n" );
 161 }
 162
 163 $reader = new BackupReader();
 164 if( isset( $options['quiet'] ) ) {
 165         $reader->reporting = false;
 166 }
 167 if( isset( $options['report'] ) ) {
 168         $reader->reportingInterval = intval( $options['report'] );
 169 }
 170 if( isset( $options['dry-run'] ) ) {
 171         $reader->dryRun = true;
 172 }
 173 if( isset( $options['debug'] ) ) {
 174         $reader->debug = true;
 175 }
 176 if( isset( $options['uploads'] ) ) {
 177         $reader->uploads = true; // experimental!
 178 }
 179
 180 if( isset( $args[0] ) ) {
 181         $result = $reader->importFromFile( $args[0] );
 182 } else {
 183         $result = $reader->importFromStdin();
 184 }
 185
 186 if( WikiError::isError( $result ) ) {
 187         echo $result->getMessage() . "\n";
 188 } else {
 189         echo "Done!\n";
 190         echo "You might want to run rebuildrecentchanges.php to regenerate\n";
 191         echo "the recentchanges page.\n";
 192 }
 193
 194