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