3 * Look for 'orphan' revisions hooked to pages which don't exist and
4 * 'childless' pages with no revisions.
5 * Then, kill the poor widows and orphans.
6 * Man this is depressing.
8 * Copyright © 2005 Brion Vibber <brion@pobox.com>
9 * https://www.mediawiki.org/
11 * This program is free software; you can redistribute it and/or modify
12 * it under the terms of the GNU General Public License as published by
13 * the Free Software Foundation; either version 2 of the License, or
14 * (at your option) any later version.
16 * This program is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 * GNU General Public License for more details.
21 * You should have received a copy of the GNU General Public License along
22 * with this program; if not, write to the Free Software Foundation, Inc.,
23 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
24 * http://www.gnu.org/copyleft/gpl.html
27 * @author <brion@pobox.com>
28 * @ingroup Maintenance
31 require_once __DIR__
. '/Maintenance.php';
33 use MediaWiki\MediaWikiServices
;
34 use Wikimedia\Rdbms\IMaintainableDatabase
;
37 * Maintenance script that looks for 'orphan' revisions hooked to pages which
38 * don't exist and 'childless' pages with no revisions.
40 * @ingroup Maintenance
42 class Orphans
extends Maintenance
{
43 public function __construct() {
44 parent
::__construct();
45 $this->addDescription( "Look for 'orphan' revisions hooked to pages which don't exist\n" .
46 "and 'childless' pages with no revisions\n" .
47 "Then, kill the poor widows and orphans\n" .
48 "Man this is depressing"
50 $this->addOption( 'fix', 'Actually fix broken entries' );
53 public function execute() {
54 $this->checkOrphans( $this->hasOption( 'fix' ) );
55 $this->checkSeparation( $this->hasOption( 'fix' ) );
59 * Lock the appropriate tables for the script
60 * @param IMaintainableDatabase $db
61 * @param string[] $extraTable The name of any extra tables to lock (eg: text)
63 private function lockTables( $db, $extraTable = [] ) {
64 $tbls = [ 'page', 'revision', 'redirect' ];
66 $tbls = array_merge( $tbls, $extraTable );
68 $db->lockTables( [], $tbls, __METHOD__
);
72 * Check for orphan revisions
73 * @param bool $fix Whether to fix broken revisions when found
75 private function checkOrphans( $fix ) {
76 $dbw = $this->getDB( DB_MASTER
);
77 $commentStore = CommentStore
::getStore();
80 $this->lockTables( $dbw );
83 $commentQuery = $commentStore->getJoin( 'rev_comment' );
84 $actorQuery = ActorMigration
::newMigration()->getJoin( 'rev_user' );
86 $this->output( "Checking for orphan revision table entries... "
87 . "(this may take a while on a large wiki)\n" );
88 $result = $dbw->select(
89 [ 'revision', 'page' ] +
$commentQuery['tables'] +
$actorQuery['tables'],
90 [ 'rev_id', 'rev_page', 'rev_timestamp' ] +
$commentQuery['fields'] +
$actorQuery['fields'],
91 [ 'page_id' => null ],
94 [ 'page' => [ 'LEFT JOIN', [ 'rev_page=page_id' ] ] ] +
$commentQuery['joins']
95 +
$actorQuery['joins']
97 $orphans = $result->numRows();
99 $this->output( "$orphans orphan revisions...\n" );
100 $this->output( sprintf(
101 "%10s %10s %14s %20s %s\n",
102 'rev_id', 'rev_page', 'rev_timestamp', 'rev_user_text', 'rev_comment'
105 $contLang = MediaWikiServices
::getInstance()->getContentLanguage();
106 foreach ( $result as $row ) {
107 $comment = $commentStore->getComment( 'rev_comment', $row )->text
;
108 if ( $comment !== '' ) {
109 $comment = '(' . $contLang->truncateForVisual( $comment, 40 ) . ')';
111 $rev_user_text = $contLang->truncateForVisual( $row->rev_user_text
, 20 );
112 # pad $rev_user_text to 20 characters. Note that this may
113 # yield poor results if $rev_user_text contains combining
114 # or half-width characters. Alas.
115 if ( mb_strlen( $rev_user_text ) < 20 ) {
116 $rev_user_text = str_repeat( ' ', 20 - mb_strlen( $rev_user_text ) );
118 $this->output( sprintf( "%10d %10d %14s %s %s\n",
125 $dbw->delete( 'revision', [ 'rev_id' => $row->rev_id
] );
129 $this->output( "Run again with --fix to remove these entries automatically.\n" );
132 $this->output( "No orphans! Yay!\n" );
136 $dbw->unlockTables( __METHOD__
);
141 * Check for pages where page_latest is wrong
142 * @param bool $fix Whether to fix broken entries
144 private function checkSeparation( $fix ) {
145 $dbw = $this->getDB( DB_MASTER
);
146 $page = $dbw->tableName( 'page' );
147 $revision = $dbw->tableName( 'revision' );
150 $this->lockTables( $dbw, [ 'user', 'text' ] );
153 $this->output( "\nChecking for pages whose page_latest links are incorrect... "
154 . "(this may take a while on a large wiki)\n" );
155 $result = $dbw->query( "
157 FROM $page LEFT JOIN $revision ON page_latest=rev_id
160 foreach ( $result as $row ) {
161 $result2 = $dbw->query( "
162 SELECT MAX(rev_timestamp) as max_timestamp
164 WHERE rev_page=" . (int)( $row->page_id
)
166 $row2 = $dbw->fetchObject( $result2 );
168 if ( $row->rev_timestamp
!= $row2->max_timestamp
) {
170 $this->output( sprintf( "%10s %10s %14s %14s\n",
171 'page_id', 'rev_id', 'timestamp', 'max timestamp' ) );
174 $this->output( sprintf( "%10d %10d %14s %14s\n",
178 $row2->max_timestamp
) );
181 $maxId = $dbw->selectField(
185 'rev_page' => $row->page_id
,
186 'rev_timestamp' => $row2->max_timestamp
] );
187 $this->output( "... updating to revision $maxId\n" );
188 $maxRev = Revision
::newFromId( $maxId );
189 $title = Title
::makeTitle( $row->page_namespace
, $row->page_title
);
190 $article = WikiPage
::factory( $title );
191 $article->updateRevisionOn( $dbw, $maxRev );
195 $this->output( "wtf\n" );
200 $this->output( "Found $found pages with incorrect latest revision.\n" );
202 $this->output( "No pages with incorrect latest revision. Yay!\n" );
204 if ( !$fix && $found > 0 ) {
205 $this->output( "Run again with --fix to remove these entries automatically.\n" );
209 $dbw->unlockTables( __METHOD__
);
214 $maintClass = Orphans
::class;
215 require_once RUN_MAINTENANCE_IF_MAIN
;