MediaWiki master
migrateExternallinks.php
Go to the documentation of this file.
1<?php
2
5
6// @codeCoverageIgnoreStart
7require_once __DIR__ . '/Maintenance.php';
8// @codeCoverageIgnoreEnd
9
17 public function __construct() {
18 parent::__construct();
19 $this->addDescription(
20 'Migrate externallinks data'
21 );
22 $this->addOption(
23 'sleep',
24 'Sleep time (in seconds) between every batch. Default: 0',
25 false,
26 true
27 );
28 $this->setBatchSize( 1000 );
29 }
30
32 protected function getUpdateKey() {
33 return __CLASS__;
34 }
35
37 protected function doDBUpdates() {
38 $dbw = $this->getDB( DB_PRIMARY );
39 $table = 'externallinks';
40 if ( !$dbw->fieldExists( $table, 'el_to', __METHOD__ ) ) {
41 $this->output( "Old fields don't exist. There is no need to run this script\n" );
42 return true;
43 }
44 if ( !$dbw->fieldExists( $table, 'el_to_path', __METHOD__ ) ) {
45 $this->output( "Run update.php to create the el_to_path column.\n" );
46 return false;
47 }
48
49 $this->output( "Populating el_to_domain_index and el_to_path columns\n" );
50 $updated = 0;
51
52 $highestId = $dbw->newSelectQueryBuilder()
53 ->select( 'el_id' )
54 ->from( $table )
55 ->limit( 1 )
56 ->caller( __METHOD__ )
57 ->orderBy( 'el_id', 'DESC' )
58 ->fetchResultSet()->fetchRow();
59 if ( !$highestId ) {
60 $this->output( "Page table is empty.\n" );
61 return true;
62 }
63 $highestId = $highestId[0];
64 $id = 0;
65 while ( $id <= $highestId ) {
66 $updated += $this->handleBatch( $id );
67 $id += $this->getBatchSize();
68 }
69
70 $this->output( "Completed normalization of $table, $updated rows updated.\n" );
71
72 return true;
73 }
74
75 private function handleBatch( int $lowId ): int {
76 $batchSize = $this->getBatchSize();
77 // range is inclusive, let's subtract one.
78 $highId = $lowId + $batchSize - 1;
79 $dbw = $this->getPrimaryDB();
80 $updated = 0;
81 $res = $dbw->newSelectQueryBuilder()
82 ->select( [ 'el_id', 'el_to' ] )
83 ->from( 'externallinks' )
84 ->where( [
85 'el_to_domain_index' => '',
86 $dbw->expr( 'el_id', '>=', $lowId ),
87 $dbw->expr( 'el_id', '<=', $highId ),
88 ] )
89 ->limit( $batchSize )
90 ->caller( __METHOD__ )
91 ->fetchResultSet();
92 if ( !$res->numRows() ) {
93 return $updated;
94 }
95 foreach ( $res as $row ) {
96 $url = $row->el_to;
97 $paths = LinkFilter::makeIndexes( $url );
98 if ( !$paths ) {
99 continue;
100 }
101 $dbw->newUpdateQueryBuilder()
102 ->update( 'externallinks' )
103 // just take the first one, we are not sending proto-relative to LinkFilter
104 ->set( [
105 'el_to_domain_index' => substr( $paths[0][0], 0, 255 ),
106 'el_to_path' => $paths[0][1]
107 ] )
108 ->where( [ 'el_id' => $row->el_id ] )
109 ->caller( __METHOD__ )->execute();
110
111 $updated += $dbw->affectedRows();
112 }
113 $this->output( "Updated $updated rows\n" );
114 // Sleep between batches for replication to catch up
115 $this->waitForReplication();
116 $sleep = (int)$this->getOption( 'sleep', 0 );
117 if ( $sleep > 0 ) {
118 sleep( $sleep );
119 }
120 return $updated;
121 }
122
123}
124
125// @codeCoverageIgnoreStart
126$maintClass = MigrateExternallinks::class;
127require_once RUN_MAINTENANCE_IF_MAIN;
128// @codeCoverageIgnoreEnd
Class for scripts that perform database maintenance and want to log the update in updatelog so we can...
getBatchSize()
Returns batch size.
output( $out, $channel=null)
Throw some output to the user.
addOption( $name, $description, $required=false, $withArg=false, $shortName=false, $multiOccurrence=false)
Add a parameter to the script.
getDB( $db, $groups=[], $dbDomain=false)
Returns a database to be used by current maintenance script.
waitForReplication()
Wait for replica DB servers to catch up.
getOption( $name, $default=null)
Get an option, or return the default.
addDescription( $text)
Set the description text.
const DB_PRIMARY
Definition defines.php:28