includes/filebackend/filejournal/DBFileJournal.php

   1 <?php
   2 /**
   3  * Version of FileJournal that logs to a DB table.
   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 FileJournal
  22  */
  23
  24 use MediaWiki\MediaWikiServices;
  25 use Wikimedia\Rdbms\IDatabase;
  26 use Wikimedia\Rdbms\DBError;
  27 use Wikimedia\Timestamp\ConvertibleTimestamp;
  28
  29 /**
  30  * Version of FileJournal that logs to a DB table
  31  * @since 1.20
  32  */
  33 class DBFileJournal extends FileJournal {
  34         /** @var IDatabase */
  35         protected $dbw;
  36         /** @var string */
  37         protected $domain;
  38
  39         /**
  40          * Construct a new instance from configuration. Do not call directly, use FileJournal::factory.
  41          *
  42          * @param array $config Includes:
  43          *   domain: database domain ID of the wiki
  44          */
  45         public function __construct( array $config ) {
  46                 parent::__construct( $config );
  47
  48                 $this->domain = $config['domain'] ?? $config['wiki']; // b/c
  49         }
  50
  51         /**
  52          * @see FileJournal::logChangeBatch()
  53          * @param array $entries
  54          * @param string $batchId
  55          * @return StatusValue
  56          */
  57         protected function doLogChangeBatch( array $entries, $batchId ) {
  58                 $status = StatusValue::newGood();
  59
  60                 try {
  61                         $dbw = $this->getMasterDB();
  62                 } catch ( DBError $e ) {
  63                         $status->fatal( 'filejournal-fail-dbconnect', $this->backend );
  64
  65                         return $status;
  66                 }
  67
  68                 $now = ConvertibleTimestamp::time();
  69
  70                 $data = [];
  71                 foreach ( $entries as $entry ) {
  72                         $data[] = [
  73                                 'fj_batch_uuid' => $batchId,
  74                                 'fj_backend' => $this->backend,
  75                                 'fj_op' => $entry['op'],
  76                                 'fj_path' => $entry['path'],
  77                                 'fj_new_sha1' => $entry['newSha1'],
  78                                 'fj_timestamp' => $dbw->timestamp( $now )
  79                         ];
  80                 }
  81
  82                 try {
  83                         $dbw->insert( 'filejournal', $data, __METHOD__ );
  84                         // XXX Should we do this deterministically so it's testable? Maybe look at the last two
  85                         // digits of a hash of a bunch of the data?
  86                         if ( mt_rand( 0, 99 ) == 0 ) {
  87                                 // occasionally delete old logs
  88                                 $this->purgeOldLogs(); // @codeCoverageIgnore
  89                         }
  90                 } catch ( DBError $e ) {
  91                         $status->fatal( 'filejournal-fail-dbquery', $this->backend );
  92
  93                         return $status;
  94                 }
  95
  96                 return $status;
  97         }
  98
  99         /**
 100          * @see FileJournal::doGetCurrentPosition()
 101          * @return bool|mixed The value from the field, or false on failure.
 102          */
 103         protected function doGetCurrentPosition() {
 104                 $dbw = $this->getMasterDB();
 105
 106                 return $dbw->selectField( 'filejournal', 'MAX(fj_id)',
 107                         [ 'fj_backend' => $this->backend ],
 108                         __METHOD__
 109                 );
 110         }
 111
 112         /**
 113          * @see FileJournal::doGetPositionAtTime()
 114          * @param int|string $time Timestamp
 115          * @return bool|mixed The value from the field, or false on failure.
 116          */
 117         protected function doGetPositionAtTime( $time ) {
 118                 $dbw = $this->getMasterDB();
 119
 120                 $encTimestamp = $dbw->addQuotes( $dbw->timestamp( $time ) );
 121
 122                 return $dbw->selectField( 'filejournal', 'fj_id',
 123                         [ 'fj_backend' => $this->backend, "fj_timestamp <= $encTimestamp" ],
 124                         __METHOD__,
 125                         [ 'ORDER BY' => 'fj_timestamp DESC' ]
 126                 );
 127         }
 128
 129         /**
 130          * @see FileJournal::doGetChangeEntries()
 131          * @param int|null $start
 132          * @param int $limit
 133          * @return array[]
 134          */
 135         protected function doGetChangeEntries( $start, $limit ) {
 136                 $dbw = $this->getMasterDB();
 137
 138                 $res = $dbw->select( 'filejournal', '*',
 139                         [
 140                                 'fj_backend' => $this->backend,
 141                                 'fj_id >= ' . $dbw->addQuotes( (int)$start ) ], // $start may be 0
 142                         __METHOD__,
 143                         array_merge( [ 'ORDER BY' => 'fj_id ASC' ],
 144                                 $limit ? [ 'LIMIT' => $limit ] : [] )
 145                 );
 146
 147                 $entries = [];
 148                 foreach ( $res as $row ) {
 149                         $item = [];
 150                         foreach ( (array)$row as $key => $value ) {
 151                                 $item[substr( $key, 3 )] = $value; // "fj_op" => "op"
 152                         }
 153                         $entries[] = $item;
 154                 }
 155
 156                 return $entries;
 157         }
 158
 159         /**
 160          * @see FileJournal::purgeOldLogs()
 161          * @return StatusValue
 162          * @throws DBError
 163          */
 164         protected function doPurgeOldLogs() {
 165                 $status = StatusValue::newGood();
 166                 if ( $this->ttlDays <= 0 ) {
 167                         return $status; // nothing to do
 168                 }
 169
 170                 $dbw = $this->getMasterDB();
 171                 $dbCutoff = $dbw->timestamp( ConvertibleTimestamp::time() - 86400 * $this->ttlDays );
 172
 173                 $dbw->delete( 'filejournal',
 174                         [ 'fj_timestamp < ' . $dbw->addQuotes( $dbCutoff ) ],
 175                         __METHOD__
 176                 );
 177
 178                 return $status;
 179         }
 180
 181         /**
 182          * Get a master connection to the logging DB
 183          *
 184          * @return IDatabase
 185          * @throws DBError
 186          */
 187         protected function getMasterDB() {
 188                 if ( !$this->dbw ) {
 189                         // Get a separate connection in autocommit mode
 190                         $lb = MediaWikiServices::getInstance()->getDBLoadBalancerFactory()->newMainLB();
 191                         $this->dbw = $lb->getConnection( DB_MASTER, [], $this->domain );
 192                         $this->dbw->clearFlag( DBO_TRX );
 193                 }
 194
 195                 return $this->dbw;
 196         }
 197 }