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 = null;
  78                 if ( $dbr->tableExists( 'valid_tag' ) ) {
  79                         $userTags = $dbr->selectFieldValues(
  80                                 'valid_tag',
  81                                 'vt_tag',
  82                                 [],
  83                                 __METHOD__
  84                         );
  85                 }
  86
  87                 if ( empty( $userTags ) ) {
  88                         $this->output( "No user defined tags to set, moving on...\n" );
  89                         return;
  90                 }
  91
  92                 if ( $this->hasOption( 'dry-run' ) ) {
  93                         $this->output(
  94                                 'These tags will have ctd_user_defined=1 : ' . implode( ', ', $userTags ) . "\n"
  95                         );
  96                         return;
  97                 }
  98
  99                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 100
 101                 $dbw->update(
 102                         'change_tag_def',
 103                         [ 'ctd_user_defined' => 1 ],
 104                         [ 'ctd_name' => $userTags ],
 105                         __METHOD__
 106                 );
 107                 $this->lbFactory->waitForReplication();
 108                 $this->output( "Finished setting user defined tags in change_tag_def table\n" );
 109         }
 110
 111         private function updateCountTagId() {
 112                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 113
 114                 // This query can be pretty expensive, don't run it on master
 115                 $res = $dbr->select(
 116                         'change_tag',
 117                         [ 'ct_tag_id', 'hitcount' => 'count(*)' ],
 118                         [],
 119                         __METHOD__,
 120                         [ 'GROUP BY' => 'ct_tag_id' ]
 121                 );
 122
 123                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 124
 125                 foreach ( $res as $row ) {
 126                         if ( !$row->ct_tag_id ) {
 127                                 continue;
 128                         }
 129
 130                         if ( $this->hasOption( 'dry-run' ) ) {
 131                                 $this->output( 'This row will be updated: ' . implode( ', ', $row ) . "\n" );
 132                                 continue;
 133                         }
 134
 135                         $dbw->update(
 136                                 'change_tag_def',
 137                                 [ 'ctd_count' => $row->hitcount ],
 138                                 [ 'ctd_id' => $row->ct_tag_id ],
 139                                 __METHOD__
 140                         );
 141                 }
 142                 $this->lbFactory->waitForReplication();
 143         }
 144
 145         private function updateCountTag() {
 146                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 147
 148                 // This query can be pretty expensive, don't run it on master
 149                 $res = $dbr->select(
 150                         'change_tag',
 151                         [ 'ct_tag', 'hitcount' => 'count(*)' ],
 152                         [],
 153                         __METHOD__,
 154                         [ 'GROUP BY' => 'ct_tag' ]
 155                 );
 156
 157                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 158
 159                 foreach ( $res as $row ) {
 160                         // Hygiene check
 161                         if ( !$row->ct_tag ) {
 162                                 continue;
 163                         }
 164
 165                         if ( $this->hasOption( 'dry-run' ) ) {
 166                                 $this->output( 'This row will be updated: ' . $row->ct_tag . $row->hitcount . "\n" );
 167                                 continue;
 168                         }
 169
 170                         $dbw->upsert(
 171                                 'change_tag_def',
 172                                 [
 173                                         'ctd_name' => $row->ct_tag,
 174                                         'ctd_user_defined' => 0,
 175                                         'ctd_count' => $row->hitcount
 176                                 ],
 177                                 [ 'ctd_name' ],
 178                                 [ 'ctd_count' => $row->hitcount ],
 179                                 __METHOD__
 180                         );
 181                 }
 182                 $this->lbFactory->waitForReplication();
 183         }
 184
 185         private function backpopulateChangeTagId() {
 186                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 187                 $changeTagDefs = $dbr->select(
 188                         'change_tag_def',
 189                         [ 'ctd_name', 'ctd_id' ],
 190                         [],
 191                         __METHOD__,
 192                         [ 'ORDER BY' => 'ctd_id' ]
 193                 );
 194
 195                 foreach ( $changeTagDefs as $row ) {
 196                         $this->backpopulateChangeTagPerTag( $row->ctd_name, $row->ctd_id );
 197                 }
 198         }
 199
 200         private function backpopulateChangeTagPerTag( $tagName, $tagId ) {
 201                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 202                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 203                 $sleep = (int)$this->getOption( 'sleep', 0 );
 204                 $lastId = 0;
 205                 $this->output( "Starting to add ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 206                 while ( true ) {
 207                         // Given that indexes might not be there, it's better to use replica
 208                         $ids = $dbr->selectFieldValues(
 209                                 'change_tag',
 210                                 'ct_id',
 211                                 [ 'ct_tag' => $tagName, 'ct_tag_id' => null, 'ct_id > ' . $lastId ],
 212                                 __METHOD__,
 213                                 [ 'LIMIT' => $this->getBatchSize(), 'ORDER BY' => 'ct_id' ]
 214                         );
 215
 216                         if ( !$ids ) {
 217                                 break;
 218                         }
 219                         $lastId = end( $ids );
 220
 221                         if ( $this->hasOption( 'dry-run' ) ) {
 222                                 $this->output(
 223                                         "These ids will be changed to have \"{$tagId}\" as tag id: " . implode( ', ', $ids ) . "\n"
 224                                 );
 225                                 continue;
 226                         } else {
 227                                 $this->output( "Updating ct_tag_id = {$tagId} up to row ct_id = {$lastId}\n" );
 228                         }
 229
 230                         $dbw->update(
 231                                 'change_tag',
 232                                 [ 'ct_tag_id' => $tagId ],
 233                                 [ 'ct_id' => $ids ],
 234                                 __METHOD__
 235                         );
 236
 237                         $this->lbFactory->waitForReplication();
 238                         if ( $sleep > 0 ) {
 239                                 sleep( $sleep );
 240                         }
 241                 }
 242
 243                 $this->output( "Finished adding ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 244         }
 245
 246         protected function getUpdateKey() {
 247                 return __CLASS__;
 248         }
 249 }
 250
 251 $maintClass = PopulateChangeTagDef::class;
 252 require_once RUN_MAINTENANCE_IF_MAIN;