3 * Fills the rev_sha1 and ar_sha1 columns of revision
4 * and archive tables for revisions created before MW 1.19.
6 * This program is free software; you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License as published by
8 * the Free Software Foundation; either version 2 of the License, or
9 * (at your option) any later version.
11 * This program is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 * GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License along
17 * with this program; if not, write to the Free Software Foundation, Inc.,
18 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
19 * http://www.gnu.org/copyleft/gpl.html
22 * @ingroup Maintenance
25 require_once __DIR__
. '/Maintenance.php';
28 * Maintenance script that fills the rev_sha1 and ar_sha1 columns of revision
29 * and archive tables for revisions created before MW 1.19.
31 * @ingroup Maintenance
33 class PopulateRevisionSha1
extends LoggedUpdateMaintenance
{
34 public function __construct() {
35 parent
::__construct();
36 $this->addDescription( 'Populates the rev_sha1 and ar_sha1 fields' );
37 $this->setBatchSize( 200 );
40 protected function getUpdateKey() {
41 return 'populate rev_sha1';
44 protected function doDBUpdates() {
45 $db = $this->getDB( DB_MASTER
);
47 if ( !$db->tableExists( 'revision' ) ) {
48 $this->fatalError( "revision table does not exist" );
49 } elseif ( !$db->tableExists( 'archive' ) ) {
50 $this->fatalError( "archive table does not exist" );
51 } elseif ( !$db->fieldExists( 'revision', 'rev_sha1', __METHOD__
) ) {
52 $this->output( "rev_sha1 column does not exist\n\n", true );
57 $this->output( "Populating rev_sha1 column\n" );
58 $rc = $this->doSha1Updates( 'revision', 'rev_id', Revision
::getQueryInfo(), 'rev' );
60 $this->output( "Populating ar_sha1 column\n" );
61 $ac = $this->doSha1Updates( 'archive', 'ar_rev_id', Revision
::getArchiveQueryInfo(), 'ar' );
62 $this->output( "Populating ar_sha1 column legacy rows\n" );
63 $ac +
= $this->doSha1LegacyUpdates();
65 $this->output( "rev_sha1 and ar_sha1 population complete "
66 . "[$rc revision rows, $ac archive rows].\n" );
72 * @param string $table
73 * @param string $idCol
74 * @param array $queryInfo
75 * @param string $prefix
76 * @return int Rows changed
78 protected function doSha1Updates( $table, $idCol, $queryInfo, $prefix ) {
79 $db = $this->getDB( DB_MASTER
);
80 $batchSize = $this->getBatchSize();
81 $start = $db->selectField( $table, "MIN($idCol)", false, __METHOD__
);
82 $end = $db->selectField( $table, "MAX($idCol)", false, __METHOD__
);
83 if ( !$start ||
!$end ) {
84 $this->output( "...$table table seems to be empty.\n" );
91 $end +
= $batchSize - 1;
93 $blockEnd = $start +
$batchSize - 1;
94 while ( $blockEnd <= $end ) {
95 $this->output( "...doing $idCol from $blockStart to $blockEnd\n" );
96 $cond = "$idCol BETWEEN " . (int)$blockStart . " AND " . (int)$blockEnd .
97 " AND $idCol IS NOT NULL AND {$prefix}_sha1 = ''";
99 $queryInfo['tables'], $queryInfo['fields'], $cond, __METHOD__
, [], $queryInfo['joins']
102 $this->beginTransaction( $db, __METHOD__
);
103 foreach ( $res as $row ) {
104 if ( $this->upgradeRow( $row, $table, $idCol, $prefix ) ) {
108 $this->commitTransaction( $db, __METHOD__
);
110 $blockStart +
= $batchSize;
111 $blockEnd +
= $batchSize;
121 protected function doSha1LegacyUpdates() {
123 $db = $this->getDB( DB_MASTER
);
124 $arQuery = Revision
::getArchiveQueryInfo();
125 $res = $db->select( $arQuery['tables'], $arQuery['fields'],
126 [ 'ar_rev_id IS NULL', 'ar_sha1' => '' ], __METHOD__
, [], $arQuery['joins'] );
129 $this->beginTransaction( $db, __METHOD__
);
130 foreach ( $res as $row ) {
131 if ( $this->upgradeLegacyArchiveRow( $row ) ) {
134 if ( ++
$updateSize >= 100 ) {
136 $this->commitTransaction( $db, __METHOD__
);
137 $this->output( "Commited row with ar_timestamp={$row->ar_timestamp}\n" );
139 $this->beginTransaction( $db, __METHOD__
);
142 $this->commitTransaction( $db, __METHOD__
);
148 * @param stdClass $row
149 * @param string $table
150 * @param string $idCol
151 * @param string $prefix
154 protected function upgradeRow( $row, $table, $idCol, $prefix ) {
155 $db = $this->getDB( DB_MASTER
);
157 $rev = ( $table === 'archive' )
158 ? Revision
::newFromArchiveRow( $row )
159 : new Revision( $row );
160 $text = $rev->getSerializedData();
161 } catch ( Exception
$e ) {
162 $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
164 return false; // T24624?
166 if ( !is_string( $text ) ) {
167 # This should not happen, but sometimes does (T22757)
168 $this->output( "Data of revision with {$idCol}={$row->$idCol} unavailable!\n" );
173 [ "{$prefix}_sha1" => Revision
::base36Sha1( $text ) ],
174 [ $idCol => $row->$idCol ],
183 * @param stdClass $row
186 protected function upgradeLegacyArchiveRow( $row ) {
187 $db = $this->getDB( DB_MASTER
);
189 $rev = Revision
::newFromArchiveRow( $row );
190 } catch ( Exception
$e ) {
191 $this->output( "Text of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
193 return false; // T24624?
195 $text = $rev->getSerializedData();
196 if ( !is_string( $text ) ) {
197 # This should not happen, but sometimes does (T22757)
198 $this->output( "Data of revision with timestamp {$row->ar_timestamp} unavailable!\n" );
202 # Archive table as no PK, but (NS,title,time) should be near unique.
203 # Any duplicates on those should also have duplicated text anyway.
204 $db->update( 'archive',
205 [ 'ar_sha1' => Revision
::base36Sha1( $text ) ],
207 'ar_namespace' => $row->ar_namespace
,
208 'ar_title' => $row->ar_title
,
209 'ar_timestamp' => $row->ar_timestamp
,
210 'ar_len' => $row->ar_len
// extra sanity
220 $maintClass = "PopulateRevisionSha1";
221 require_once RUN_MAINTENANCE_IF_MAIN
;