maintenance/storage/moveToExternal.php

   1 <?php
   2 /**
   3  * Move revision's text to external storage
   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 ExternalStorage
  22  */
  23
  24 define( 'REPORTING_INTERVAL', 1 );
  25
  26 if ( !defined( 'MEDIAWIKI' ) ) {
  27         require_once __DIR__ . '/../commandLine.inc';
  28         require_once 'resolveStubs.php';
  29
  30         $fname = 'moveToExternal';
  31
  32         if ( !isset( $args[0] ) ) {
  33                 print "Usage: php moveToExternal.php [-s <startid>] [-e <endid>] <cluster>\n";
  34                 exit;
  35         }
  36
  37         $cluster = $args[0];
  38         $dbw = wfGetDB( DB_MASTER );
  39
  40         if ( isset( $options['e'] ) ) {
  41                 $maxID = $options['e'];
  42         } else {
  43                 $maxID = $dbw->selectField( 'text', 'MAX(old_id)', false, $fname );
  44         }
  45         $minID = isset( $options['s'] ) ? $options['s'] : 1;
  46
  47         moveToExternal( $cluster, $maxID, $minID );
  48 }
  49
  50 function moveToExternal( $cluster, $maxID, $minID = 1 ) {
  51         $fname = 'moveToExternal';
  52         $dbw = wfGetDB( DB_MASTER );
  53         $dbr = wfGetDB( DB_SLAVE );
  54
  55         $count = $maxID - $minID + 1;
  56         $blockSize = 1000;
  57         $numBlocks = ceil( $count / $blockSize );
  58         print "Moving text rows from $minID to $maxID to external storage\n";
  59         $ext = new ExternalStoreDB;
  60         $numMoved = 0;
  61
  62         for ( $block = 0; $block < $numBlocks; $block++ ) {
  63                 $blockStart = $block * $blockSize + $minID;
  64                 $blockEnd = $blockStart + $blockSize - 1;
  65
  66                 if ( !( $block % REPORTING_INTERVAL ) ) {
  67                         print "oldid=$blockStart, moved=$numMoved\n";
  68                         wfWaitForSlaves();
  69                 }
  70
  71                 $res = $dbr->select( 'text', [ 'old_id', 'old_flags', 'old_text' ],
  72                         [
  73                                 "old_id BETWEEN $blockStart AND $blockEnd",
  74                                 'old_flags NOT ' . $dbr->buildLike( $dbr->anyString(), 'external', $dbr->anyString() ),
  75                         ], $fname );
  76                 foreach ( $res as $row ) {
  77                         # Resolve stubs
  78                         $text = $row->old_text;
  79                         $id = $row->old_id;
  80                         if ( $row->old_flags === '' ) {
  81                                 $flags = 'external';
  82                         } else {
  83                                 $flags = "{$row->old_flags},external";
  84                         }
  85
  86                         if ( strpos( $flags, 'object' ) !== false ) {
  87                                 $obj = unserialize( $text );
  88                                 $className = strtolower( get_class( $obj ) );
  89                                 if ( $className == 'historyblobstub' ) {
  90                                         # resolveStub( $id, $row->old_text, $row->old_flags );
  91                                         # $numStubs++;
  92                                         continue;
  93                                 } elseif ( $className == 'historyblobcurstub' ) {
  94                                         $text = gzdeflate( $obj->getText() );
  95                                         $flags = 'utf-8,gzip,external';
  96                                 } elseif ( $className == 'concatenatedgziphistoryblob' ) {
  97                                         // Do nothing
  98                                 } else {
  99                                         print "Warning: unrecognised object class \"$className\"\n";
 100                                         continue;
 101                                 }
 102                         } else {
 103                                 $className = false;
 104                         }
 105
 106                         if ( strlen( $text ) < 100 && $className === false ) {
 107                                 // Don't move tiny revisions
 108                                 continue;
 109                         }
 110
 111                         # print "Storing "  . strlen( $text ) . " bytes to $url\n";
 112                         # print "old_id=$id\n";
 113
 114                         $url = $ext->store( $cluster, $text );
 115                         if ( !$url ) {
 116                                 print "Error writing to external storage\n";
 117                                 exit;
 118                         }
 119                         $dbw->update( 'text',
 120                                 [ 'old_flags' => $flags, 'old_text' => $url ],
 121                                 [ 'old_id' => $id ], $fname );
 122                         $numMoved++;
 123                 }
 124         }
 125 }