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
  29 /**
  30  * @ingroup Maintenance
  31  * @todo port to Maintenance class
  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 __construct() {
  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( $filename ) {
 119                 if ( preg_match( '/\.gz$/', $filename ) ) {
 120                         $filename = 'compress.zlib://' . $filename;
 121                 }
 122                 elseif ( preg_match( '/\.bz2$/', $filename ) ) {
 123                         $filename = 'compress.bzip2://' . $filename;
 124                 }
 125                 elseif ( preg_match( '/\.7z$/', $filename ) ) {
 126                         $filename = 'mediawiki.compress.7z://' . $filename;
 127                 }
 128
 129                 $file = fopen( $filename, 'rt' );
 130                 return $this->importFromHandle( $file );
 131         }
 132
 133         function importFromStdin() {
 134                 $file = fopen( 'php://stdin', 'rt' );
 135                 if( posix_isatty( $file ) ) {
 136                         $this->showHelp();
 137                         exit();
 138                 }
 139                 return $this->importFromHandle( $file );
 140         }
 141
 142         function importFromHandle( $handle ) {
 143                 $this->startTime = wfTime();
 144
 145                 $source = new ImportStreamSource( $handle );
 146                 $importer = new WikiImporter( $source );
 147
 148                 $importer->setDebug( $this->debug );
 149                 $importer->setPageCallback( array( &$this, 'reportPage' ) );
 150                 $this->importCallback =  $importer->setRevisionCallback(
 151                         array( &$this, 'handleRevision' ) );
 152                 $this->uploadCallback = $importer->setUploadCallback(
 153                         array( &$this, 'handleUpload' ) );
 154                 $this->logItemCallback = $importer->setLogItemCallback(
 155                         array( &$this, 'handleLogItem' ) );
 156
 157                 if ( $this->dryRun ) {
 158                         $importer->setPageOutCallback( null );
 159                 }
 160
 161                 return $importer->doImport();
 162         }
 163
 164         function showHelp() {
 165                 $gz = in_array('compress.zlib', stream_get_wrappers()) ? 'ok' : '(disabled; requires PHP zlib module)';
 166                 $bz2 = in_array('compress.bzip2', stream_get_wrappers()) ? 'ok' : '(disabled; requires PHP bzip2 module)';
 167                 echo "This script reads pages from an XML file as produced from Special:Export\n";
 168                 echo "or dumpBackup.php, and saves them into the current wiki.\n";
 169                 echo "\n";
 170                 echo "Note that for very large data sets, importDump.php may be slow; there are\n";
 171                 echo "alternate methods which can be much faster for full site restoration:\n";
 172                 echo "http://www.mediawiki.org/wiki/Manual:Importing_XML_dumps\n";
 173                 echo "\n";
 174                 echo "Usage: php importDump.php [<options>] [<file>]\n";
 175                 echo "If no file is listed, input may be piped from stdin.\n";
 176                 echo "\n";
 177                 echo "Options:\n";
 178                 echo "  --quiet    Don't dump status reports to stderr.\n";
 179                 echo "  --report=n Report position and speed after every n pages processed.\n";
 180                 echo "  --dry-run  Parse dump without actually importing pages.\n";
 181                 echo "  --debug    Output extra verbose debug information\n";
 182                 echo "  --uploads  Process file upload data if included (experimental)\n";
 183                 echo "\n";
 184                 echo "Compressed XML files may be read directly:\n";
 185                 echo "  .gz $gz\n";
 186                 echo "  .bz2 $bz2\n";
 187                 echo "  .7z (if 7za executable is in PATH)\n";
 188                 echo "\n";
 189         }
 190 }
 191
 192 if ( wfReadOnly() ) {
 193         wfDie( "Wiki is in read-only mode; you'll need to disable it for import to work.\n" );
 194 }
 195
 196 $reader = new BackupReader();
 197 if ( isset( $options['quiet'] ) ) {
 198         $reader->reporting = false;
 199 }
 200 if ( isset( $options['report'] ) ) {
 201         $reader->reportingInterval = intval( $options['report'] );
 202 }
 203 if ( isset( $options['dry-run'] ) ) {
 204         $reader->dryRun = true;
 205 }
 206 if ( isset( $options['debug'] ) ) {
 207         $reader->debug = true;
 208 }
 209 if ( isset( $options['uploads'] ) ) {
 210         $reader->uploads = true; // experimental!
 211 }
 212
 213 if ( isset( $options['help'] ) ) {
 214         $reader->showHelp();
 215         exit();
 216 } elseif ( isset( $args[0] ) ) {
 217         $result = $reader->importFromFile( $args[0] );
 218 } else {
 219         $result = $reader->importFromStdin();
 220 }
 221
 222 echo "Done!\n";
 223 echo "You might want to run rebuildrecentchanges.php to regenerate\n";
 224 echo "the recentchanges page.\n";