maintenance/purgeList.php

   1 <?php
   2 /**
   3  * Send purge requests for listed pages to squid
   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  * @file
  21  * @ingroup Maintenance
  22  */
  23
  24 require_once( __DIR__ . '/Maintenance.php' );
  25
  26 /**
  27  * Maintenance script that sends purge requests for listed pages to squid.
  28  *
  29  * @ingroup Maintenance
  30  */
  31 class PurgeList extends Maintenance {
  32         public function __construct() {
  33                 parent::__construct();
  34                 $this->mDescription = "Send purge requests for listed pages to squid";
  35                 $this->addOption( 'purge', 'Whether to update page_touched.' , false, false );
  36                 $this->addOption( 'namespace', 'Namespace number', false, true );
  37                 $this->setBatchSize( 100 );
  38         }
  39
  40         public function execute() {
  41                 if( $this->hasOption( 'namespace' ) ) {
  42                         $this->purgeNamespace();
  43                 } else {
  44                         $this->purgeList();
  45                 }
  46                 $this->output( "Done!\n" );
  47         }
  48
  49         /** Purge URL coming from stdin */
  50         private function purgeList() {
  51                 $stdin = $this->getStdin();
  52                 $urls = array();
  53
  54                 while ( !feof( $stdin ) ) {
  55                         $page = trim( fgets( $stdin ) );
  56                         if ( preg_match( '%^https?://%', $page ) ) {
  57                                 $urls[] = $page;
  58                         } elseif ( $page !== '' ) {
  59                                 $title = Title::newFromText( $page );
  60                                 if ( $title ) {
  61                                         $url = $title->getInternalUrl();
  62                                         $this->output( "$url\n" );
  63                                         $urls[] = $url;
  64                                         if ( $this->getOption( 'purge' ) ) {
  65                                                 $title->invalidateCache();
  66                                         }
  67                                 } else {
  68                                         $this->output( "(Invalid title '$page')\n" );
  69                                 }
  70                         }
  71                 }
  72                 $this->sendPurgeRequest( $urls );
  73         }
  74
  75         /** Purge a namespace given by --namespace */
  76         private function purgeNamespace() {
  77                 $dbr = wfGetDB( DB_SLAVE );
  78                 $ns = $dbr->addQuotes( $this->getOption( 'namespace') );
  79
  80                 $result = $dbr->select(
  81                         array( 'page' ),
  82                         array( 'page_namespace', 'page_title' ),
  83                         array( "page_namespace = $ns" ),
  84                         __METHOD__,
  85                         array( 'ORDER BY' => 'page_id' )
  86                 );
  87
  88                 $start   = 0;
  89                 $end = $dbr->numRows( $result );
  90                 $this->output( "Will purge $end pages from namespace $ns\n" );
  91
  92                 # Do remaining chunk
  93                 $end += $this->mBatchSize - 1;
  94                 $blockStart = $start;
  95                 $blockEnd = $start + $this->mBatchSize - 1;
  96
  97                 while( $blockEnd <= $end ) {
  98                         # Select pages we will purge:
  99                         $result = $dbr->select(
 100                                 array( 'page' ),
 101                                 array( 'page_namespace', 'page_title' ),
 102                                 array( "page_namespace = $ns" ),
 103                                 __METHOD__,
 104                                 array( # conditions
 105                                         'ORDER BY' => 'page_id',
 106                                         'LIMIT'    => $this->mBatchSize,
 107                                         'OFFSET'   => $blockStart,
 108                                 )
 109                         );
 110                         # Initialize/reset URLs to be purged
 111                         $urls = array();
 112                         foreach( $result as $row ) {
 113                                 $title = Title::makeTitle( $row->page_namespace, $row->page_title );
 114                                 $url = $title->getInternalUrl();
 115                                 $urls[] = $url;
 116                         }
 117
 118                         $this->sendPurgeRequest( $urls );
 119
 120                         $blockStart += $this->mBatchSize;
 121                         $blockEnd   += $this->mBatchSize;
 122                 }
 123         }
 124
 125         /**
 126          * Helper to purge an array of $urls
 127          * @param $urls array List of URLS to purge from squids
 128          */
 129         private function sendPurgeRequest( $urls ) {
 130                 $this->output( "Purging " . count( $urls ). " urls\n" );
 131                 $u = new SquidUpdate( $urls );
 132                 $u->doUpdate();
 133         }
 134
 135 }
 136
 137 $maintClass = "PurgeList";
 138 require_once( RUN_MAINTENANCE_IF_MAIN );