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',
  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         public function execute() {
  46                 global $wgChangeTagsSchemaMigrationStage;
  47                 if ( $wgChangeTagsSchemaMigrationStage === MIGRATION_OLD ) {
  48                         // Return "success", but don't flag it as done so the next run will retry
  49                         $this->output( '... Not run, $wgChangeTagsSchemaMigrationStage === MIGRATION_OLD' . "\n" );
  50                         return true;
  51                 }
  52                 return parent::execute();
  53         }
  54
  55         protected function doDBUpdates() {
  56                 $this->lbFactory = MediaWiki\MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
  57                 $this->setBatchSize( $this->getOption( 'batch-size', $this->getBatchSize() ) );
  58
  59                 if ( $this->lbFactory->getMainLB()->getConnection( DB_REPLICA )->fieldExists(
  60                                 'change_tag',
  61                                 'ct_tag',
  62                                 __METHOD__
  63                         )
  64                 ) {
  65                         if ( $this->hasOption( 'set-user-tags-only' ) ) {
  66                                 $this->setUserDefinedTags();
  67                                 return true;
  68                         }
  69                         if ( !$this->hasOption( 'populate-only' ) ) {
  70                                 $this->updateCountTag();
  71                         }
  72                         $this->backpopulateChangeTagId();
  73                         $this->setUserDefinedTags();
  74                 } else {
  75                         $this->updateCountTagId();
  76                 }
  77
  78                 // TODO: Implement
  79                 // $this->cleanZeroCountRows();
  80
  81                 return true;
  82         }
  83
  84         private function setUserDefinedTags() {
  85                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
  86
  87                 $userTags = $dbr->selectFieldValues(
  88                         'valid_tag',
  89                         'vt_tag',
  90                         [],
  91                         __METHOD__
  92                 );
  93
  94                 if ( empty( $userTags ) ) {
  95                         $this->output( "No user defined tags to set, moving on...\n" );
  96                         return;
  97                 }
  98
  99                 if ( $this->hasOption( 'dry-run' ) ) {
 100                         $this->output(
 101                                 'These tags will have ctd_user_defined=1 : ' . implode( ', ', $userTags ) . "\n"
 102                         );
 103                         return;
 104                 }
 105
 106                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 107
 108                 $dbw->update(
 109                         'change_tag_def',
 110                         [ 'ctd_user_defined' => 1 ],
 111                         [ 'ctd_name' => $userTags ],
 112                         __METHOD__
 113                 );
 114                 $this->lbFactory->waitForReplication();
 115                 $this->output( "Finished setting user defined tags in change_tag_def table\n" );
 116         }
 117
 118         private function updateCountTagId() {
 119                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 120
 121                 // This query can be pretty expensive, don't run it on master
 122                 $res = $dbr->select(
 123                         'change_tag',
 124                         [ 'ct_tag_id', 'hitcount' => 'count(*)' ],
 125                         [],
 126                         __METHOD__,
 127                         [ 'GROUP BY' => 'ct_tag_id' ]
 128                 );
 129
 130                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 131
 132                 foreach ( $res as $row ) {
 133                         if ( !$row->ct_tag_id ) {
 134                                 continue;
 135                         }
 136
 137                         if ( $this->hasOption( 'dry-run' ) ) {
 138                                 $this->output( 'This row will be updated: ' . implode( ', ', $row ) . "\n" );
 139                                 continue;
 140                         }
 141
 142                         $dbw->update(
 143                                 'change_tag_def',
 144                                 [ 'ctd_count' => $row->hitcount ],
 145                                 [ 'ctd_id' => $row->ct_tag_id ],
 146                                 __METHOD__
 147                         );
 148                 }
 149                 $this->lbFactory->waitForReplication();
 150         }
 151
 152         private function updateCountTag() {
 153                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 154
 155                 // This query can be pretty expensive, don't run it on master
 156                 $res = $dbr->select(
 157                         'change_tag',
 158                         [ 'ct_tag', 'hitcount' => 'count(*)' ],
 159                         [],
 160                         __METHOD__,
 161                         [ 'GROUP BY' => 'ct_tag' ]
 162                 );
 163
 164                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 165
 166                 foreach ( $res as $row ) {
 167                         // Hygiene check
 168                         if ( !$row->ct_tag ) {
 169                                 continue;
 170                         }
 171
 172                         if ( $this->hasOption( 'dry-run' ) ) {
 173                                 $this->output( 'This row will be updated: ' . $row->ct_tag . $row->hitcount . "\n" );
 174                                 continue;
 175                         }
 176
 177                         $dbw->upsert(
 178                                 'change_tag_def',
 179                                 [
 180                                         'ctd_name' => $row->ct_tag,
 181                                         'ctd_user_defined' => 0,
 182                                         'ctd_count' => $row->hitcount
 183                                 ],
 184                                 [ 'ctd_name' ],
 185                                 [ 'ctd_count' => $row->hitcount ],
 186                                 __METHOD__
 187                         );
 188                 }
 189                 $this->lbFactory->waitForReplication();
 190         }
 191
 192         private function backpopulateChangeTagId() {
 193                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 194                 $changeTagDefs = $dbr->select(
 195                         'change_tag_def',
 196                         [ 'ctd_name', 'ctd_id' ],
 197                         [],
 198                         __METHOD__,
 199                         [ 'ORDER BY' => 'ctd_id' ]
 200                 );
 201
 202                 foreach ( $changeTagDefs as $row ) {
 203                         $this->backpopulateChangeTagPerTag( $row->ctd_name, $row->ctd_id );
 204                 }
 205         }
 206
 207         private function backpopulateChangeTagPerTag( $tagName, $tagId ) {
 208                 $dbr = $this->lbFactory->getMainLB()->getConnection( DB_REPLICA );
 209                 $dbw = $this->lbFactory->getMainLB()->getConnection( DB_MASTER );
 210                 $sleep = (int)$this->getOption( 'sleep', 10 );
 211                 $lastId = 0;
 212                 $this->output( "Starting to add ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 213                 while ( true ) {
 214                         // Given that indexes might not be there, it's better to use replica
 215                         $ids = $dbr->selectFieldValues(
 216                                 'change_tag',
 217                                 'ct_id',
 218                                 [ 'ct_tag' => $tagName, 'ct_tag_id' => null, 'ct_id > ' . $lastId ],
 219                                 __METHOD__,
 220                                 [ 'LIMIT' => $this->getBatchSize(), 'ORDER BY' => 'ct_id' ]
 221                         );
 222
 223                         if ( !$ids ) {
 224                                 break;
 225                         }
 226                         $lastId = end( $ids );
 227
 228                         if ( $this->hasOption( 'dry-run' ) ) {
 229                                 $this->output(
 230                                         "These ids will be changed to have \"{$tagId}\" as tag id: " . implode( ', ', $ids ) . "\n"
 231                                 );
 232                                 continue;
 233                         } else {
 234                                 $this->output( "Updating ct_tag_id = {$tagId} up to row ct_id = {$lastId}\n" );
 235                         }
 236
 237                         $dbw->update(
 238                                 'change_tag',
 239                                 [ 'ct_tag_id' => $tagId ],
 240                                 [ 'ct_id' => $ids ],
 241                                 __METHOD__
 242                         );
 243
 244                         $this->lbFactory->waitForReplication();
 245                         if ( $sleep > 0 ) {
 246                                 sleep( $sleep );
 247                         }
 248                 }
 249
 250                 $this->output( "Finished adding ct_tag_id = {$tagId} for ct_tag = {$tagName}\n" );
 251         }
 252
 253         protected function getUpdateKey() {
 254                 return __CLASS__;
 255         }
 256 }
 257
 258 $maintClass = PopulateChangeTagDef::class;
 259 require_once RUN_MAINTENANCE_IF_MAIN;