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 require_once __DIR__ . '/Maintenance.php';
  20
  21 /**
  22  * Populate and improve accuracy of change_tag_def statistics.
  23  *
  24  * @ingroup Maintenance
  25  */
  26 class PopulateChangeTagDef extends LoggedUpdateMaintenance {
  27         /** @var Wikimedia\Rdbms\ILBFactory */
  28         protected $lbFactory;
  29
  30         public function __construct() {
  31                 parent::__construct();
  32                 $this->addDescription( 'Populate and improve accuracy of change_tag_def statistics' );
  33                 $this->addOption( 'dry-run', 'Print debug info instead of actually deleting' );
  34                 $this->setBatchSize( 1000 );
  35                 $this->addOption(
  36                         'sleep',
  37                         'Sleep time (in seconds) between every batch, defaults to zero',
  38                         false,
  39                         true
  40                 );
  41                 $this->addOption( 'populate-only', 'Do not update change_tag_def table' );
  42                 $this->addOption( 'set-user-tags-only', 'Only update ctd_user_defined from valid_tag table' );
  43         }
  44
  45         protected function doDBUpdates() {
  46                 $this->lbFactory = MediaWiki\MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
  47                 $this->setBatchSize( $this->getOption( 'batch-size', $this->getBatchSize() ) );
  48
  49                 if ( $this->lbFactory->getMainLB()->getConnection( DB_REPLICA )->fieldExists(
  50                                 'change_tag',
  51                                 'ct_tag',
  52                                 __METHOD__
  53                         )
  54                 ) {
  55                         if ( $this->hasOption( 'set-user-tags-only' ) ) {
  56                                 $this->setUserDefinedTags();
  57                                 return true;
  58                         }
  59                         if ( !$this->hasOption( 'populate-only' ) ) {
  60                                 $this->updateCountTag();
  61                         }
  62                         $this->backpopulateChangeTagId();
  63                         $this->setUserDefinedTags();
  64                 } else {
  65                         $this->updateCountTagId();
  66                 }
  67
  68                 // TODO: Implement
  69                 // $this->cleanZeroCountRows();
  70
  71                 return true;
  72         }
  73
  74         private function setUserDefinedTags() {
  75                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
  76
  77                 $userTags = $dbr->selectFieldValues(
  78                         'valid_tag',
  79                         'vt_tag',
  80                         [],
  81                         __METHOD__
  82                 );
  83
  84                 if ( empty( $userTags ) ) {
  85                         $this->output( "No user defined tags to set, moving on...\n" );
  86                         return;
  87                 }
  88
  89                 if ( $this->hasOption( 'dry-run' ) ) {
  90                         $this->output(
  91                                 'These tags will have ctd_user_defined=1 : ' . implode( ', ', $userTags ) . "\n"
  92                         );
  93                         return;
  94                 }
  95
  96                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
  97
  98                 $dbw->update(
  99                         'change_tag_def',
 100                         [ 'ctd_user_defined' => 1 ],
 101                         [ 'ctd_name' => $userTags ],
 102                         __METHOD__
 103                 );
 104                 $this->lbFactory->waitForReplication();
 105                 $this->output( "Finished setting user defined tags in change_tag_def table\n" );
 106         }
 107
 108         private function updateCountTagId() {
 109                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 110
 111                 // This query can be pretty expensive, don't run it on master
 112                 $res = $dbr->select(
 113                         'change_tag',
 114                         [ 'ct_tag_id', 'hitcount' => 'count(*)' ],
 115                         [],
 116                         __METHOD__,
 117                         [ 'GROUP BY' => 'ct_tag_id' ]
 118                 );
 119
 120                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 121
 122                 foreach ( $res as $row ) {
 123                         if ( !$row->ct_tag_id ) {
 124                                 continue;
 125                         }
 126
 127                         if ( $this->hasOption( 'dry-run' ) ) {
 128                                 $this->output( 'This row will be updated: ' . implode( ', ', $row ) . "\n" );
 129                                 continue;
 130                         }
 131
 132                         $dbw->update(
 133                                 'change_tag_def',
 134                                 [ 'ctd_count' => $row->hitcount ],
 135                                 [ 'ctd_id' => $row->ct_tag_id ],
 136                                 __METHOD__
 137                         );
 138                 }
 139                 $this->lbFactory->waitForReplication();
 140         }
 141
 142         private function updateCountTag() {
 143                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 144
 145                 // This query can be pretty expensive, don't run it on master
 146                 $res = $dbr->select(
 147                         'change_tag',
 148                         [ 'ct_tag', 'hitcount' => 'count(*)' ],
 149                         [],
 150                         __METHOD__,
 151                         [ 'GROUP BY' => 'ct_tag' ]
 152                 );
 153
 154                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 155
 156                 foreach ( $res as $row ) {
 157                         // Hygiene check
 158                         if ( !$row->ct_tag ) {
 159                                 continue;
 160                         }
 161
 162                         if ( $this->hasOption( 'dry-run' ) ) {
 163                                 $this->output( 'This row will be updated: ' . $row->ct_tag . $row->hitcount . "\n" );
 164                                 continue;
 165                         }
 166
 167                         $dbw->upsert(
 168                                 'change_tag_def',
 169                                 [
 170                                         'ctd_name' => $row->ct_tag,
 171                                         'ctd_user_defined' => 0,
 172                                         'ctd_count' => $row->hitcount
 173                                 ],
 174                                 [ 'ctd_name' ],
 175                                 [ 'ctd_count' => $row->hitcount ],
 176                                 __METHOD__
 177                         );
 178                 }
 179                 $this->lbFactory->waitForReplication();
 180         }
 181
 182         private function backpopulateChangeTagId() {
 183                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 184                 $changeTagDefs = $dbr->select(
 185                         'change_tag_def',
 186                         [ 'ctd_name', 'ctd_id' ],
 187                         [],
 188                         __METHOD__,
 189                         [ 'ORDER BY' => 'ctd_id' ]
 190                 );
 191
 192                 foreach ( $changeTagDefs as $row ) {
 193                         $this->backpopulateChangeTagPerTag( $row->ctd_name, $row->ctd_id );
 194                 }
 195         }
 196
 197         private function backpopulateChangeTagPerTag( $tagName, $tagId ) {
 198                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 199                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 200                 $sleep = (int)$this->getOption( 'sleep', 0 );
 201                 $lastId = 0;
 202                 $this->output( "Starting to add ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 203                 while ( true ) {
 204                         // Given that indexes might not be there, it's better to use replica
 205                         $ids = $dbr->selectFieldValues(
 206                                 'change_tag',
 207                                 'ct_id',
 208                                 [ 'ct_tag' => $tagName, 'ct_tag_id' => null, 'ct_id > ' . $lastId ],
 209                                 __METHOD__,
 210                                 [ 'LIMIT' => $this->getBatchSize(), 'ORDER BY' => 'ct_id' ]
 211                         );
 212
 213                         if ( !$ids ) {
 214                                 break;
 215                         }
 216                         $lastId = end( $ids );
 217
 218                         if ( $this->hasOption( 'dry-run' ) ) {
 219                                 $this->output(
 220                                         "These ids will be changed to have \"{$tagId}\" as tag id: " . implode( ', ', $ids ) . "\n"
 221                                 );
 222                                 continue;
 223                         } else {
 224                                 $this->output( "Updating ct_tag_id = {$tagId} up to row ct_id = {$lastId}\n" );
 225                         }
 226
 227                         $dbw->update(
 228                                 'change_tag',
 229                                 [ 'ct_tag_id' => $tagId ],
 230                                 [ 'ct_id' => $ids ],
 231                                 __METHOD__
 232                         );
 233
 234                         $this->lbFactory->waitForReplication();
 235                         if ( $sleep > 0 ) {
 236                                 sleep( $sleep );
 237                         }
 238                 }
 239
 240                 $this->output( "Finished adding ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 241         }
 242
 243         protected function getUpdateKey() {
 244                 return __CLASS__;
 245         }
 246 }
 247
 248 $maintClass = PopulateChangeTagDef::class;
 249 require_once RUN_MAINTENANCE_IF_MAIN;