Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
0.00% |
0 / 123 |
|
0.00% |
0 / 2 |
CRAP | |
0.00% |
0 / 1 |
| MigrateESRefToAflTable | |
0.00% |
0 / 121 |
|
0.00% |
0 / 2 |
552 | |
0.00% |
0 / 1 |
| __construct | |
0.00% |
0 / 23 |
|
0.00% |
0 / 1 |
2 | |||
| execute | |
0.00% |
0 / 98 |
|
0.00% |
0 / 1 |
506 | |||
| 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 | * @file |
| 19 | * @ingroup Maintenance ExternalStorage |
| 20 | */ |
| 21 | |
| 22 | namespace MediaWiki\Extension\AbuseFilter\Maintenance; |
| 23 | |
| 24 | // @codeCoverageIgnoreStart |
| 25 | $IP = getenv( 'MW_INSTALL_PATH' ); |
| 26 | if ( $IP === false ) { |
| 27 | $IP = __DIR__ . '/../../..'; |
| 28 | } |
| 29 | require_once "$IP/maintenance/Maintenance.php"; |
| 30 | // @codeCoverageIgnoreEnd |
| 31 | |
| 32 | use InvalidArgumentException; |
| 33 | use MediaWiki\MainConfigNames; |
| 34 | use MediaWiki\Maintenance\Maintenance; |
| 35 | use MediaWiki\Storage\SqlBlobStore; |
| 36 | use Wikimedia\Rdbms\SelectQueryBuilder; |
| 37 | |
| 38 | /** |
| 39 | * Migrates references of the text table which again references external storage in the abuse_filter_log |
| 40 | * table to the abuse_filter_log table directly referencing external storage and getting rid of the row |
| 41 | * in the text table. |
| 42 | */ |
| 43 | class MigrateESRefToAflTable extends Maintenance { |
| 44 | |
| 45 | public function __construct() { |
| 46 | parent::__construct(); |
| 47 | |
| 48 | $this->addOption( 'start', 'start afl_id', false, true, 's' ); |
| 49 | $this->addOption( 'end', 'end afl_id', false, true, 'e' ); |
| 50 | $this->addOption( 'dry-run', 'Don\'t modify any rows' ); |
| 51 | $this->addOption( |
| 52 | 'sleep', |
| 53 | 'Sleep time (in seconds) between every batch. Default: 0', |
| 54 | false, |
| 55 | true |
| 56 | ); |
| 57 | $this->addOption( |
| 58 | 'deletedump', |
| 59 | 'Filename to dump the list of text table rows which can later be deleted', |
| 60 | true, |
| 61 | true |
| 62 | ); |
| 63 | $this->addOption( |
| 64 | 'dump', |
| 65 | 'Filename to dump tt: -> es:DB references to.', |
| 66 | false, |
| 67 | true |
| 68 | ); |
| 69 | |
| 70 | $this->setBatchSize( 100 ); |
| 71 | } |
| 72 | |
| 73 | public function execute() { |
| 74 | if ( !in_array( 'DB', $this->getConfig()->get( MainConfigNames::ExternalStores ) ) ) { |
| 75 | $this->fatalError( 'This maintenance script is for use with external storage.' ); |
| 76 | } |
| 77 | |
| 78 | $dbw = $this->getPrimaryDB(); |
| 79 | $batchSize = $this->getBatchSize(); |
| 80 | $dryRun = $this->getOption( 'dry-run', false ); |
| 81 | $sleep = (float)$this->getOption( 'sleep', 0 ); |
| 82 | |
| 83 | $maxID = $this->getOption( 'end' ); |
| 84 | if ( $maxID === null ) { |
| 85 | $maxID = $dbw->newSelectQueryBuilder() |
| 86 | ->select( 'afl_id' ) |
| 87 | ->from( 'abuse_filter_log' ) |
| 88 | ->orderBy( 'afl_id', SelectQueryBuilder::SORT_DESC ) |
| 89 | ->limit( 1 ) |
| 90 | ->caller( __METHOD__ ) |
| 91 | ->fetchField(); |
| 92 | } |
| 93 | $maxID = (int)$maxID; |
| 94 | $minID = (int)$this->getOption( 'start', 1 ); |
| 95 | |
| 96 | $diff = $maxID - $minID + 1; |
| 97 | |
| 98 | $dump = $this->getOption( 'dump', false ); |
| 99 | if ( $dump ) { |
| 100 | $dumpfile = fopen( $dump, 'a' ); |
| 101 | if ( !$dumpfile ) { |
| 102 | $this->fatalError( "Failed to open file {$dump}." ); |
| 103 | } |
| 104 | } else { |
| 105 | $dumpfile = false; |
| 106 | } |
| 107 | |
| 108 | $deletedump = $this->getOption( 'deletedump' ); |
| 109 | $deletedumpfile = fopen( $deletedump, 'a' ); |
| 110 | if ( !$deletedumpfile ) { |
| 111 | $this->fatalError( "Failed to open file {$deletedump}." ); |
| 112 | } |
| 113 | |
| 114 | while ( true ) { |
| 115 | $res = $dbw->newSelectQueryBuilder() |
| 116 | ->select( [ 'afl_id', 'afl_var_dump' ] ) |
| 117 | ->from( 'abuse_filter_log' ) |
| 118 | ->conds( [ |
| 119 | $dbw->expr( 'afl_id', '>=', $minID ), |
| 120 | $dbw->expr( 'afl_id', '<', $minID + $batchSize ), |
| 121 | ] ) |
| 122 | ->caller( __METHOD__ ) |
| 123 | ->fetchResultSet(); |
| 124 | |
| 125 | foreach ( $res as $row ) { |
| 126 | try { |
| 127 | [ $schema, $id, ] = SqlBlobStore::splitBlobAddress( $row->afl_var_dump ); |
| 128 | } catch ( InvalidArgumentException $ex ) { |
| 129 | $this->output( $ex->getMessage() . ". Use findBadBlobs.php to remedy.\n" ); |
| 130 | continue; |
| 131 | } |
| 132 | |
| 133 | // Skip blobs which already reference external storage directly |
| 134 | if ( $schema === 'es' ) { |
| 135 | continue; |
| 136 | } |
| 137 | |
| 138 | // Skip bad blobs |
| 139 | if ( $schema !== 'tt' ) { |
| 140 | $this->output( "abuse filter log id {$row->afl_id} has special stuff: {$row->afl_var_dump}\n" ); |
| 141 | continue; |
| 142 | } |
| 143 | |
| 144 | $oldId = intval( $id ); |
| 145 | |
| 146 | if ( !$oldId ) { |
| 147 | $this->output( "Malformed text_id: $oldId\n" ); |
| 148 | continue; |
| 149 | } |
| 150 | |
| 151 | $textRow = $dbw->newSelectQueryBuilder() |
| 152 | ->select( [ 'old_text', 'old_flags' ] ) |
| 153 | ->from( 'text' ) |
| 154 | ->where( [ 'old_id' => $oldId ] ) |
| 155 | ->caller( __METHOD__ ) |
| 156 | ->fetchRow(); |
| 157 | |
| 158 | if ( !$textRow ) { |
| 159 | $this->output( "Text row for a blob of metadata of {$row->afl_id} is missing.\n" ); |
| 160 | continue; |
| 161 | } |
| 162 | |
| 163 | $flags = SqlBlobStore::explodeFlags( $textRow->old_flags ); |
| 164 | |
| 165 | if ( !in_array( 'external', $flags ) ) { |
| 166 | $this->output( "old_id {$oldId} is not external.\n" ); |
| 167 | continue; |
| 168 | } |
| 169 | |
| 170 | $newFlags = implode( ',', array_filter( |
| 171 | $flags, |
| 172 | static function ( $v ) { |
| 173 | return $v !== 'external'; |
| 174 | } |
| 175 | ) ); |
| 176 | $newBlobAddress = 'es:' . $textRow->old_text . '?flags=' . $newFlags; |
| 177 | |
| 178 | if ( !$dryRun ) { |
| 179 | $dbw->newUpdateQueryBuilder() |
| 180 | ->update( 'abuse_filter_log' ) |
| 181 | ->set( [ 'afl_var_dump' => $newBlobAddress ] ) |
| 182 | ->where( [ 'afl_id' => $row->afl_id ] ) |
| 183 | ->caller( __METHOD__ ) |
| 184 | ->execute(); |
| 185 | |
| 186 | if ( $deletedumpfile ) { |
| 187 | fwrite( $deletedumpfile, $row->afl_var_dump . "\n" ); |
| 188 | } |
| 189 | |
| 190 | if ( $dumpfile ) { |
| 191 | fwrite( $dumpfile, $row->afl_var_dump . " => " . $newBlobAddress . ";\n" ); |
| 192 | } |
| 193 | } else { |
| 194 | $this->output( "DRY-RUN: Would set blob address for {$row->afl_id} to " |
| 195 | . "{$newBlobAddress} and mark text row {$oldId} for deletion.\n" ); |
| 196 | } |
| 197 | } |
| 198 | |
| 199 | $this->waitForReplication(); |
| 200 | if ( $sleep > 0 ) { |
| 201 | if ( $sleep >= 1 ) { |
| 202 | sleep( (int)$sleep ); |
| 203 | } else { |
| 204 | usleep( (int)( $sleep * 1000000 ) ); |
| 205 | } |
| 206 | } |
| 207 | |
| 208 | $this->output( "Processed {$res->numRows()} rows out of $diff.\n" ); |
| 209 | |
| 210 | $minID += $batchSize; |
| 211 | if ( $minID > $maxID ) { |
| 212 | break; |
| 213 | } |
| 214 | } |
| 215 | |
| 216 | if ( $dumpfile ) { |
| 217 | fclose( $dumpfile ); |
| 218 | } |
| 219 | |
| 220 | if ( $deletedumpfile ) { |
| 221 | fclose( $deletedumpfile ); |
| 222 | } |
| 223 | } |
| 224 | } |
| 225 | |
| 226 | $maintClass = MigrateESRefToAflTable::class; |
| 227 | require_once RUN_MAINTENANCE_IF_MAIN; |