maintenance/cleanupTable.inc

   1 <?php
   2 /**
   3  * Generic table cleanup class. Already subclasses maintenance
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @defgroup Wikimedia
  21  * @ingroup Maintenance
  22  */
  23
  24 require_once( dirname(__FILE__) . '/Maintenance.php' );
  25
  26 abstract class TableCleanup extends Maintenance {
  27         protected $targetTable = 'page';
  28         protected $dryrun = false;
  29         protected $maxLag = 10; # if slaves are lagged more than 10 secs, wait
  30
  31         public function __construct() {
  32                 parent::__construct();
  33                 $this->addOption( 'dry-run', 'Perform a dry run' );
  34         }
  35
  36         public function execute() {
  37                 global $wgUser;
  38                 $wgUser->setName( 'Conversion script' );
  39                 $this->dryrun = $this->hasOption( 'dry-run' );
  40                 if( $this->dryrun ) {
  41                         $this->output( "Checking for bad titles...\n" );
  42                 } else {
  43                         $this->output( "Checking and fixing bad titles...\n" );
  44                 }
  45                 $this->runTable( $this->targetTable,
  46                         '', //'WHERE page_namespace=0',
  47                         array( $this, 'processPage' ) );
  48         }
  49
  50         protected function init( $count, $table ) {
  51                 $this->processed = 0;
  52                 $this->updated = 0;
  53                 $this->count = $count;
  54                 $this->startTime = wfTime();
  55                 $this->table = $table;
  56         }
  57
  58         protected function progress( $updated ) {
  59                 $this->updated += $updated;
  60                 $this->processed++;
  61                 if( $this->processed % 100 != 0 ) {
  62                         return;
  63                 }
  64                 $portion = $this->processed / $this->count;
  65                 $updateRate = $this->updated / $this->processed;
  66
  67                 $now = wfTime();
  68                 $delta = $now - $this->startTime;
  69                 $estimatedTotalTime = $delta / $portion;
  70                 $eta = $this->startTime + $estimatedTotalTime;
  71
  72                 $this->output(
  73                         sprintf( "%s %s: %6.2f%% done on %s; ETA %s [%d/%d] %.2f/sec <%.2f%% updated>\n",
  74                                 wfWikiID(),
  75                                 wfTimestamp( TS_DB, intval( $now ) ),
  76                                 $portion * 100.0,
  77                                 $this->table,
  78                                 wfTimestamp( TS_DB, intval( $eta ) ),
  79                                 $this->processed,
  80                                 $this->count,
  81                                 $this->processed / $delta,
  82                                 $updateRate * 100.0
  83                         )
  84                 );
  85                 flush();
  86         }
  87
  88         protected function runTable( $table, $where, $callback ) {
  89                 $dbw = wfGetDB( DB_MASTER );
  90
  91                 // Unbuffered queries, avoids OOM
  92                 $dbw->bufferResults( false );
  93
  94                 $count = $dbw->selectField( $table, 'count(*)', '', __METHOD__ );
  95                 $this->init( $count, $table );
  96                 $this->output( "Processing $table..." );
  97
  98                 $tableName = $dbw->tableName( $table );
  99                 $sql = "SELECT * FROM $tableName $where";
 100                 $result = $dbw->query( $sql, __METHOD__ );
 101
 102                 foreach( $result as $row ) {
 103                         call_user_func( $callback, $row );
 104                 }
 105
 106                 $this->output( "Finished $table... $this->updated of $this->processed rows updated\n" );
 107
 108                 $result->free();
 109
 110                 $dbw->bufferResults( true );
 111         }
 112
 113         protected function hexChar( $matches ) {
 114                 return sprintf( "\\x%02x", ord( $matches[1] ) );
 115         }
 116
 117         abstract protected function processPage( $row );
 118 }