3 * Find all revisions by logged out users and copy the rev_id,
4 * rev_timestamp, and a hex representation of rev_user_text to the
5 * new ip_changes table. This table is used to efficiently query for
6 * contributions within an IP range.
8 * This program is free software; you can redistribute it and/or modify
9 * it under the terms of the GNU General Public License as published by
10 * the Free Software Foundation; either version 2 of the License, or
11 * (at your option) any later version.
13 * This program is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 * GNU General Public License for more details.
18 * You should have received a copy of the GNU General Public License along
19 * with this program; if not, write to the Free Software Foundation, Inc.,
20 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
21 * http://www.gnu.org/copyleft/gpl.html
24 * @ingroup Maintenance
27 require_once __DIR__
. '/Maintenance.php';
29 use MediaWiki\MediaWikiServices
;
32 * Maintenance script that will find all rows in the revision table where
33 * rev_user = 0 (user is an IP), and copy relevant fields to ip_changes so
34 * that historical data will be available when querying for IP ranges.
36 * @ingroup Maintenance
38 class PopulateIpChanges
extends LoggedUpdateMaintenance
{
39 public function __construct() {
40 parent
::__construct();
42 $this->addDescription( <<<TEXT
43 This script will find all rows in the revision table where the user is an IP,
44 and copy relevant fields to the ip_changes table. This backfilled data will
45 then be available when querying for IP ranges at Special:Contributions.
48 $this->addOption( 'rev-id', 'The rev_id to start copying from. Default: 0', false, true );
51 'The rev_id to stop at. Default: result of MAX(rev_id)',
57 'Wait this many milliseconds after copying each batch of revisions. Default: 0',
61 $this->addOption( 'force', 'Run regardless of whether the database says it\'s been run already' );
64 public function doDBUpdates() {
65 $dbw = $this->getDB( DB_MASTER
);
67 if ( !$dbw->tableExists( 'ip_changes' ) ) {
68 $this->fatalError( 'ip_changes table does not exist' );
71 $lbFactory = MediaWikiServices
::getInstance()->getDBLoadBalancerFactory();
72 $dbr = $this->getDB( DB_REPLICA
, [ 'vslow' ] );
73 $throttle = intval( $this->getOption( 'throttle', 0 ) );
74 $maxRevId = intval( $this->getOption( 'max-rev-id', 0 ) );
75 $start = $this->getOption( 'rev-id', 0 );
78 : $dbw->selectField( 'revision', 'MAX(rev_id)', false, __METHOD__
);
83 $this->output( "Copying IP revisions to ip_changes, from rev_id $start to rev_id $end\n" );
85 $actorMigration = ActorMigration
::newMigration();
86 $actorQuery = $actorMigration->getJoin( 'rev_user' );
87 $revUserIsAnon = $actorMigration->isAnon( $actorQuery['fields']['rev_user'] );
89 while ( $blockStart <= $end ) {
90 $blockEnd = min( $blockStart +
$this->getBatchSize(), $end );
92 [ 'revision' ] +
$actorQuery['tables'],
93 [ 'rev_id', 'rev_timestamp', 'rev_user_text' => $actorQuery['fields']['rev_user_text'] ],
94 [ "rev_id BETWEEN " . (int)$blockStart . " AND " . (int)$blockEnd, $revUserIsAnon ],
100 $numRows = $rows->numRows();
102 if ( !$rows ||
$numRows === 0 ) {
103 $blockStart = $blockEnd +
1;
107 $this->output( "...checking $numRows revisions for IP edits that need copying, " .
108 "between rev_ids $blockStart and $blockEnd\n" );
111 foreach ( $rows as $row ) {
112 // Make sure this is really an IP, e.g. not maintenance user or imported revision.
113 if ( IP
::isValid( $row->rev_user_text
) ) {
115 'ipc_rev_id' => $row->rev_id
,
116 'ipc_rev_timestamp' => $row->rev_timestamp
,
117 'ipc_hex' => IP
::toHex( $row->rev_user_text
),
125 $dbw->insert( 'ip_changes', $insertRows, __METHOD__
, 'IGNORE' );
127 $inserted +
= $dbw->affectedRows();
130 $lbFactory->waitForReplication();
131 usleep( $throttle * 1000 );
133 $blockStart = $blockEnd +
1;
136 $this->output( "Attempted to insert $attempted IP revisions, $inserted actually done.\n" );
141 protected function getUpdateKey() {
142 return 'populate ip_changes';
146 $maintClass = PopulateIpChanges
::class;
147 require_once RUN_MAINTENANCE_IF_MAIN
;