maintenance/populateChangeTagDef.php

   1 <?php
   2 /**
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License along
  14  * with this program; if not, write to the Free Software Foundation, Inc.,
  15  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  16  * http://www.gnu.org/copyleft/gpl.html
  17  */
  18
  19 use MediaWiki\Maintenance\LoggedUpdateMaintenance;
  20
  21 // @codeCoverageIgnoreStart
  22 require_once __DIR__ . '/Maintenance.php';
  23 // @codeCoverageIgnoreEnd
  24
  25 /**
  26  * Populate and improve accuracy of change_tag_def statistics.
  27  *
  28  * @ingroup Maintenance
  29  */
  30 class PopulateChangeTagDef extends LoggedUpdateMaintenance {
  31         public function __construct() {
  32                 parent::__construct();
  33                 $this->addDescription( 'Populate and improve accuracy of change_tag_def statistics' );
  34                 $this->addOption( 'dry-run', 'Print debug info instead of actually deleting' );
  35                 $this->setBatchSize( 1000 );
  36                 $this->addOption(
  37                         'sleep',
  38                         'Sleep time (in seconds) between every batch, defaults to zero',
  39                         false,
  40                         true
  41                 );
  42                 $this->addOption( 'populate-only', 'Do not update change_tag_def table' );
  43                 $this->addOption( 'set-user-tags-only', 'Only update ctd_user_defined from valid_tag table' );
  44         }
  45
  46         protected function doDBUpdates() {
  47                 $this->setBatchSize( $this->getOption( 'batch-size', $this->getBatchSize() ) );
  48
  49                 $dbw = $this->getDB( DB_PRIMARY );
  50                 if ( $dbw->fieldExists(
  51                                 'change_tag',
  52                                 'ct_tag',
  53                                 __METHOD__
  54                         )
  55                 ) {
  56                         if ( $this->hasOption( 'set-user-tags-only' ) ) {
  57                                 $this->setUserDefinedTags();
  58                                 return true;
  59                         }
  60                         if ( !$this->hasOption( 'populate-only' ) ) {
  61                                 $this->updateCountTag();
  62                         }
  63                         $this->backpopulateChangeTagId();
  64                         $this->setUserDefinedTags();
  65                 } else {
  66                         $this->updateCountTagId();
  67                 }
  68
  69                 // TODO: Implement
  70                 // $this->cleanZeroCountRows();
  71
  72                 return true;
  73         }
  74
  75         private function setUserDefinedTags() {
  76                 $dbw = $this->getDB( DB_PRIMARY );
  77
  78                 $userTags = null;
  79                 if ( $dbw->tableExists( 'valid_tag', __METHOD__ ) ) {
  80                         $userTags = $dbw->newSelectQueryBuilder()
  81                                 ->select( 'vt_tag' )
  82                                 ->from( 'valid_tag' )
  83                                 ->caller( __METHOD__ )->fetchFieldValues();
  84                 }
  85
  86                 if ( !$userTags ) {
  87                         $this->output( "No user defined tags to set, moving on...\n" );
  88                         return;
  89                 }
  90
  91                 if ( $this->hasOption( 'dry-run' ) ) {
  92                         $this->output(
  93                                 'These tags will have ctd_user_defined=1 : ' . implode( ', ', $userTags ) . "\n"
  94                         );
  95                         return;
  96                 }
  97
  98                 $dbw->newUpdateQueryBuilder()
  99                         ->update( 'change_tag_def' )
 100                         ->set( [ 'ctd_user_defined' => 1 ] )
 101                         ->where( [ 'ctd_name' => $userTags ] )
 102                         ->caller( __METHOD__ )
 103                         ->execute();
 104                 $this->waitForReplication();
 105                 $this->output( "Finished setting user defined tags in change_tag_def table\n" );
 106         }
 107
 108         private function updateCountTagId() {
 109                 $dbr = $this->getReplicaDB();
 110
 111                 // This query can be pretty expensive, don't run it on master
 112                 $res = $dbr->newSelectQueryBuilder()
 113                         ->select( [ 'ct_tag_id', 'hitcount' => 'count(*)' ] )
 114                         ->from( 'change_tag' )
 115                         ->groupBy( 'ct_tag_id' )
 116                         ->caller( __METHOD__ )->fetchResultSet();
 117
 118                 $dbw = $this->getPrimaryDB();
 119
 120                 foreach ( $res as $row ) {
 121                         if ( !$row->ct_tag_id ) {
 122                                 continue;
 123                         }
 124
 125                         if ( $this->hasOption( 'dry-run' ) ) {
 126                                 $this->output( 'This row will be updated: ' . implode( ', ', $row ) . "\n" );
 127                                 continue;
 128                         }
 129
 130                         $dbw->newUpdateQueryBuilder()
 131                                 ->update( 'change_tag_def' )
 132                                 ->set( [ 'ctd_count' => $row->hitcount ] )
 133                                 ->where( [ 'ctd_id' => $row->ct_tag_id ] )
 134                                 ->caller( __METHOD__ )
 135                                 ->execute();
 136                 }
 137                 $this->waitForReplication();
 138         }
 139
 140         private function updateCountTag() {
 141                 $dbr = $this->getReplicaDB();
 142
 143                 // This query can be pretty expensive, don't run it on master
 144                 $res = $dbr->newSelectQueryBuilder()
 145                         ->select( [ 'ct_tag', 'hitcount' => 'count(*)' ] )
 146                         ->from( 'change_tag' )
 147                         ->groupBy( 'ct_tag' )
 148                         ->caller( __METHOD__ )->fetchResultSet();
 149
 150                 $dbw = $this->getPrimaryDB();
 151
 152                 foreach ( $res as $row ) {
 153                         // Hygiene check
 154                         if ( !$row->ct_tag ) {
 155                                 continue;
 156                         }
 157
 158                         if ( $this->hasOption( 'dry-run' ) ) {
 159                                 $this->output( 'This row will be updated: ' . $row->ct_tag . $row->hitcount . "\n" );
 160                                 continue;
 161                         }
 162                         $dbw->newInsertQueryBuilder()
 163                                 ->insertInto( 'change_tag_def' )
 164                                 ->row( [
 165                                         'ctd_name' => $row->ct_tag,
 166                                         'ctd_user_defined' => 0,
 167                                         'ctd_count' => $row->hitcount
 168                                 ] )
 169                                 ->onDuplicateKeyUpdate()
 170                                 ->uniqueIndexFields( [ 'ctd_name' ] )
 171                                 ->set( [ 'ctd_count' => $row->hitcount ] )
 172                                 ->caller( __METHOD__ )->execute();
 173                 }
 174                 $this->waitForReplication();
 175         }
 176
 177         private function backpopulateChangeTagId() {
 178                 $dbr = $this->getReplicaDB();
 179                 $changeTagDefs = $dbr->newSelectQueryBuilder()
 180                         ->select( [ 'ctd_name', 'ctd_id' ] )
 181                         ->from( 'change_tag_def' )
 182                         ->orderBy( 'ctd_id' )
 183                         ->caller( __METHOD__ )->fetchResultSet();
 184
 185                 foreach ( $changeTagDefs as $row ) {
 186                         $this->backpopulateChangeTagPerTag( $row->ctd_name, $row->ctd_id );
 187                 }
 188         }
 189
 190         private function backpopulateChangeTagPerTag( $tagName, $tagId ) {
 191                 $dbr = $this->getReplicaDB();
 192                 $dbw = $this->getPrimaryDB();
 193                 $sleep = (int)$this->getOption( 'sleep', 0 );
 194                 $lastId = 0;
 195                 $this->output( "Starting to add ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 196                 while ( true ) {
 197                         // Given that indexes might not be there, it's better to use replica
 198                         $ids = $dbr->newSelectQueryBuilder()
 199                                 ->select( 'ct_id' )
 200                                 ->from( 'change_tag' )
 201                                 ->where( [ 'ct_tag' => $tagName, 'ct_tag_id' => null, $dbr->expr( 'ct_id', '>', $lastId ) ] )
 202                                 ->orderBy( 'ct_id' )
 203                                 ->limit( $this->getBatchSize() )
 204                                 ->caller( __METHOD__ )->fetchFieldValues();
 205
 206                         if ( !$ids ) {
 207                                 break;
 208                         }
 209                         $lastId = end( $ids );
 210
 211                         if ( $this->hasOption( 'dry-run' ) ) {
 212                                 $this->output(
 213                                         "These ids will be changed to have \"{$tagId}\" as tag id: " . implode( ', ', $ids ) . "\n"
 214                                 );
 215                                 continue;
 216                         } else {
 217                                 $this->output( "Updating ct_tag_id = {$tagId} up to row ct_id = {$lastId}\n" );
 218                         }
 219
 220                         $dbw->newUpdateQueryBuilder()
 221                                 ->update( 'change_tag' )
 222                                 ->set( [ 'ct_tag_id' => $tagId ] )
 223                                 ->where( [ 'ct_id' => $ids ] )
 224                                 ->caller( __METHOD__ )
 225                                 ->execute();
 226
 227                         $this->waitForReplication();
 228                         if ( $sleep > 0 ) {
 229                                 sleep( $sleep );
 230                         }
 231                 }
 232
 233                 $this->output( "Finished adding ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 234         }
 235
 236         protected function getUpdateKey() {
 237                 return __CLASS__;
 238         }
 239 }
 240
 241 // @codeCoverageIgnoreStart
 242 $maintClass = PopulateChangeTagDef::class;
 243 require_once RUN_MAINTENANCE_IF_MAIN;
 244 // @codeCoverageIgnoreEnd