1<?php
2/**
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.
7 *
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.
12 *
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.
17 *
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
22 *
23 * @file
24 * @ingroup Maintenance
25 */
26
27require_once __DIR__ . '/Maintenance.php';
28
29use MediaWiki\MediaWikiServices;
30use Wikimedia\IPUtils;
31
32/**
33 * Maintenance script that will find all rows in the revision table where
34 * rev_user = 0 (user is an IP), and copy relevant fields to ip_changes so
35 * that historical data will be available when querying for IP ranges.
36 *
37 * @ingroup Maintenance
38 */
39class PopulateIpChanges extends LoggedUpdateMaintenance {
40	public function __construct() {
41		parent::__construct();
42
43		$this->addDescription( <<<TEXT
44This script will find all rows in the revision table where the user is an IP,
45and copy relevant fields to the ip_changes table. This backfilled data will
46then be available when querying for IP ranges at Special:Contributions.
47TEXT
48		);
49		$this->addOption( 'rev-id', 'The rev_id to start copying from. Default: 0', false, true );
50		$this->addOption(
51			'max-rev-id',
52			'The rev_id to stop at. Default: result of MAX(rev_id)',
53			false,
54			true
55		);
56		$this->addOption(
57			'throttle',
58			'Wait this many milliseconds after copying each batch of revisions. Default: 0',
59			false,
60			true
61		);
62		$this->addOption( 'force', 'Run regardless of whether the database says it\'s been run already' );
63	}
64
65	public function doDBUpdates() {
66		$dbw = $this->getDB( DB_PRIMARY );
67
68		if ( !$dbw->tableExists( 'ip_changes', __METHOD__ ) ) {
69			$this->fatalError( 'ip_changes table does not exist' );
70		}
71
72		$lbFactory = MediaWikiServices::getInstance()->getDBLoadBalancerFactory();
73		$dbr = $this->getDB( DB_REPLICA, [ 'vslow' ] );
74		$throttle = intval( $this->getOption( 'throttle', 0 ) );
75		$maxRevId = intval( $this->getOption( 'max-rev-id', 0 ) );
76		$start = $this->getOption( 'rev-id', 0 );
77		$end = $maxRevId > 0
78			? $maxRevId
79			: $dbw->selectField( 'revision', 'MAX(rev_id)', '', __METHOD__ );
80
81		if ( empty( $end ) ) {
82			$this->output( "No revisions found, aborting.\n" );
83			return true;
84		}
85
86		$blockStart = $start;
87		$attempted = 0;
88		$inserted = 0;
89
90		$this->output( "Copying IP revisions to ip_changes, from rev_id $start to rev_id $end\n" );
91
92		$actorMigration = ActorMigration::newMigration();
93		$actorQuery = $actorMigration->getJoin( 'rev_user' );
94		$revUserIsAnon = $actorMigration->isAnon( $actorQuery['fields']['rev_user'] );
95
96		while ( $blockStart <= $end ) {
97			$blockEnd = min( $blockStart + $this->getBatchSize(), $end );
98			$rows = $dbr->select(
99				[ 'revision' ] + $actorQuery['tables'],
100				[ 'rev_id', 'rev_timestamp', 'rev_user_text' => $actorQuery['fields']['rev_user_text'] ],
101				[ "rev_id BETWEEN " . (int)$blockStart . " AND " . (int)$blockEnd, $revUserIsAnon ],
102				__METHOD__,
103				[],
104				$actorQuery['joins']
105			);
106
107			$numRows = $rows->numRows();
108
109			if ( !$rows || $numRows === 0 ) {
110				$blockStart = $blockEnd + 1;
111				continue;
112			}
113
114			$this->output( "...checking $numRows revisions for IP edits that need copying, " .
115				"between rev_ids $blockStart and $blockEnd\n" );
116
117			$insertRows = [];
118			foreach ( $rows as $row ) {
119				// Make sure this is really an IP, e.g. not maintenance user or imported revision.
120				if ( IPUtils::isValid( $row->rev_user_text ) ) {
121					$insertRows[] = [
122						'ipc_rev_id' => $row->rev_id,
123						'ipc_rev_timestamp' => $row->rev_timestamp,
124						'ipc_hex' => IPUtils::toHex( $row->rev_user_text ),
125					];
126
127					$attempted++;
128				}
129			}
130
131			if ( $insertRows ) {
132				$dbw->insert( 'ip_changes', $insertRows, __METHOD__, [ 'IGNORE' ] );
133
134				$inserted += $dbw->affectedRows();
135			}
136
137			$lbFactory->waitForReplication();
138			usleep( $throttle * 1000 );
139
140			$blockStart = $blockEnd + 1;
141		}
142
143		$this->output( "Attempted to insert $attempted IP revisions, $inserted actually done.\n" );
144
145		return true;
146	}
147
148	protected function getUpdateKey() {
149		return 'populate ip_changes';
150	}
151}
152
153$maintClass = PopulateIpChanges::class;
154require_once RUN_MAINTENANCE_IF_MAIN;
155