maintenance/migrateArchiveText.php

   1 <?php
   2 /**
   3  * Migrate archive.ar_text and ar_flags to modern 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
  22  */
  23
  24 use MediaWiki\MediaWikiServices;
  25
  26 require_once __DIR__ . '/Maintenance.php';
  27
  28 /**
  29  * Maintenance script that migrates archive.ar_text and ar_flags to text storage
  30  *
  31  * @ingroup Maintenance
  32  * @since 1.31
  33  */
  34 class MigrateArchiveText extends LoggedUpdateMaintenance {
  35         public function __construct() {
  36                 parent::__construct();
  37                 $this->addDescription(
  38                         'Migrates content from pre-1.5 ar_text and ar_flags columns to text storage'
  39                 );
  40                 $this->addOption(
  41                         'replace-missing',
  42                         "For rows with missing or unloadable data, throw away whatever is there and\n"
  43                         . "mark them as \"error\" in the database."
  44                 );
  45         }
  46
  47         /**
  48          * Sets whether a run of this maintenance script has the force parameter set
  49          * @param bool $forced
  50          */
  51         public function setForce( $forced = true ) {
  52                 $this->mOptions['force'] = $forced;
  53         }
  54
  55         protected function getUpdateKey() {
  56                 return __CLASS__;
  57         }
  58
  59         protected function doDBUpdates() {
  60                 $replaceMissing = $this->hasOption( 'replace-missing' );
  61                 $defaultExternalStore = $this->getConfig()->get( 'DefaultExternalStore' );
  62                 // @phan-suppress-next-line PhanAccessMethodInternal
  63                 $blobStore = MediaWikiServices::getInstance()
  64                         ->getBlobStoreFactory()
  65                         ->newSqlBlobStore();
  66                 $batchSize = $this->getBatchSize();
  67
  68                 $dbr = $this->getDB( DB_REPLICA, [ 'vslow' ] );
  69                 $dbw = $this->getDB( DB_MASTER );
  70                 if ( !$dbr->fieldExists( 'archive', 'ar_text', __METHOD__ ) ||
  71                         !$dbw->fieldExists( 'archive', 'ar_text', __METHOD__ )
  72                 ) {
  73                         $this->output( "No ar_text field, so nothing to migrate.\n" );
  74                         return true;
  75                 }
  76
  77                 $this->output( "Migrating ar_text to modern storage...\n" );
  78                 $last = 0;
  79                 $count = 0;
  80                 $errors = 0;
  81                 while ( true ) {
  82                         $res = $dbr->select(
  83                                 'archive',
  84                                 [ 'ar_id', 'ar_text', 'ar_flags' ],
  85                                 [
  86                                         'ar_text_id' => null,
  87                                         "ar_id > $last",
  88                                 ],
  89                                 __METHOD__,
  90                                 [ 'LIMIT' => $batchSize, 'ORDER BY' => [ 'ar_id' ] ]
  91                         );
  92                         $numRows = $res->numRows();
  93
  94                         foreach ( $res as $row ) {
  95                                 $last = $row->ar_id;
  96
  97                                 // Recompress the text (and store in external storage, if
  98                                 // applicable) if it's not already in external storage.
  99                                 $arFlags = explode( ',', $row->ar_flags );
 100                                 if ( !in_array( 'external', $arFlags, true ) ) {
 101                                         $data = $blobStore->decompressData( $row->ar_text, $arFlags );
 102                                         if ( $data !== false ) {
 103                                                 $flags = Revision::compressRevisionText( $data );
 104
 105                                                 if ( $defaultExternalStore ) {
 106                                                         $data = ExternalStore::insertToDefault( $data );
 107                                                         if ( $flags ) {
 108                                                                 $flags .= ',';
 109                                                         }
 110                                                         $flags .= 'external';
 111                                                 }
 112                                         } elseif ( $replaceMissing ) {
 113                                                 $this->error( "Replacing missing data for row ar_id=$row->ar_id" );
 114                                                 $data = 'Missing data in migrateArchiveText.php on ' . date( 'c' );
 115                                                 $flags = 'error';
 116                                         } else {
 117                                                 $this->error( "No data for row ar_id=$row->ar_id" );
 118                                                 $errors++;
 119                                                 continue;
 120                                         }
 121                                 } else {
 122                                         $flags = $row->ar_flags;
 123                                         $data = $row->ar_text;
 124                                 }
 125
 126                                 $this->beginTransaction( $dbw, __METHOD__ );
 127                                 $dbw->insert(
 128                                         'text',
 129                                         [ 'old_text' => $data, 'old_flags' => $flags ],
 130                                         __METHOD__
 131                                 );
 132                                 $id = $dbw->insertId();
 133                                 $dbw->update(
 134                                         'archive',
 135                                         [ 'ar_text_id' => $id, 'ar_text' => '', 'ar_flags' => '' ],
 136                                         [ 'ar_id' => $row->ar_id, 'ar_text_id' => null ],
 137                                         __METHOD__
 138                                 );
 139                                 $count += $dbw->affectedRows();
 140                                 $this->commitTransaction( $dbw, __METHOD__ );
 141                         }
 142
 143                         if ( $numRows < $batchSize ) {
 144                                 // We must have reached the end
 145                                 break;
 146                         }
 147
 148                         $this->output( "... $last\n" );
 149                         // $this->commitTransaction() already waited for replication; no need to re-wait here
 150                 }
 151
 152                 $this->output( "Completed ar_text migration, $count rows updated, $errors missing data.\n" );
 153                 if ( $errors ) {
 154                         $this->output( "Run with --replace-missing to overwrite missing data with an error message.\n" );
 155                 }
 156
 157                 return $errors === 0;
 158         }
 159 }
 160
 161 $maintClass = MigrateArchiveText::class;
 162 require_once RUN_MAINTENANCE_IF_MAIN;