maintenance/refreshLinks.php

   1 <?php
   2 /**
   3  * This program is free software; you can redistribute it and/or modify
   4  * it under the terms of the GNU General Public License as published by
   5  * the Free Software Foundation; either version 2 of the License, or
   6  * (at your option) any later version.
   7  *
   8  * This program is distributed in the hope that it will be useful,
   9  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  10  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  11  * GNU General Public License for more details.
  12  *
  13  * You should have received a copy of the GNU General Public License along
  14  * with this program; if not, write to the Free Software Foundation, Inc.,
  15  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  16  * http://www.gnu.org/copyleft/gpl.html
  17  *
  18  * @ingroup Maintenance
  19  */
  20
  21 require_once( dirname(__FILE__) . '/Maintenance.php' );
  22
  23 class RefreshLinks extends Maintenance {
  24         public function __construct() {
  25                 parent::__construct();
  26                 $this->mDescription = "Refresh link tables";
  27                 $this->addOption( 'dfn-only', 'Delete links from nonexistent articles only' );
  28                 $this->addOption( 'new-only', 'Only affect articles with just a single edit' );
  29                 $this->addOption( 'redirects-only', 'Only fix redirects, not all links' );
  30                 $this->addOption( 'old-redirects-only', 'Only fix redirects with no redirect table entry' );
  31                 $this->addOption( 'm', 'Maximum replication lag', false, true );
  32                 $this->addOption( 'e', 'Last page id to refresh', false, true );
  33                 $this->addArg( 'start', 'Page_id to start from, default 1', false );
  34                 $this->setBatchSize( 100 );
  35         }
  36
  37         public function execute() {
  38                 if( !$this->hasOption( 'dfn-only' ) ) {
  39                         $start = $this->getArg( 0, 1 );
  40                         $new = $this->getOption( 'new-only', false );
  41                         $max = $this->getOption( 'm', false );
  42                         $end = $this->getOption( 'e', 0 );
  43                         $redir = $this->getOption( 'redirects-only', false );
  44                         $oldRedir = $this->getOption( 'old-redirects-only', false );
  45                         $this->doRefreshLinks( $start, $new, $max, $end, $redir, $oldRedir );
  46                 }
  47                 $this->deleteLinksFromNonexistent( $max, $this->mBatchSize );
  48         }
  49
  50         /**
  51          * Do the actual link refreshing.
  52          * @param $start int Page_id to start from
  53          * @param $newOnly bool Only do pages with 1 edit
  54          * @param $maxLag int Max DB replication lag
  55          * @param $end int Page_id to stop at
  56          * @param $redirectsOnly bool Only fix redirects
  57          * @param $oldRedirectsOnly bool Only fix redirects without redirect entries
  58          */
  59         private function doRefreshLinks( $start, $newOnly = false, $maxLag = false,
  60                                                 $end = 0, $redirectsOnly = false, $oldRedirectsOnly = false ) {
  61                 global $wgUser, $wgParser, $wgUseTidy;
  62
  63                 $reportingInterval = 100;
  64                 $dbr = wfGetDB( DB_SLAVE );
  65                 $start = intval( $start );
  66
  67                 # Don't generate TeX PNGs (lack of a sensible current directory causes errors anyway)
  68                 $wgUser->setOption('math', MW_MATH_SOURCE);
  69
  70                 # Don't generate extension images (e.g. Timeline)
  71                 if( method_exists( $wgParser, "clearTagHooks" ) ) {
  72                         $wgParser->clearTagHooks();
  73                 }
  74
  75                 # Don't use HTML tidy
  76                 $wgUseTidy = false;
  77
  78                 $what = $redirectsOnly ? "redirects" : "links";
  79
  80                 if( $oldRedirectsOnly ) {
  81                         # This entire code path is cut-and-pasted from below.  Hurrah.
  82                         $res = $dbr->query(
  83                                 "SELECT page_id ".
  84                                 "FROM page ".
  85                                 "LEFT JOIN redirect ON page_id=rd_from ".
  86                                 "WHERE page_is_redirect=1 AND rd_from IS NULL AND ".
  87                                 ($end == 0 ? "page_id >= $start"
  88                                                    : "page_id BETWEEN $start AND $end"),
  89                                 __METHOD__
  90                         );
  91                         $num = $dbr->numRows( $res );
  92                         $this->output( "Refreshing $num old redirects from $start...\n" );
  93
  94                         foreach( $res as $row ) {
  95                                 if ( !( ++$i % $reportingInterval ) ) {
  96                                         $this->output( "$i\n" );
  97                                         wfWaitForSlaves( $maxLag );
  98                                 }
  99                                 $this->fixRedirect( $row->page_id );
 100                         }
 101                 } elseif( $newOnly ) {
 102                         $this->output( "Refreshing $what from " );
 103                         $res = $dbr->select( 'page',
 104                                 array( 'page_id' ),
 105                                 array(
 106                                         'page_is_new' => 1,
 107                                         "page_id >= $start" ),
 108                                 __METHOD__
 109                         );
 110                         $num = $dbr->numRows( $res );
 111                         $this->output( "$num new articles...\n" );
 112
 113                         $i = 0;
 114                         foreach ( $res as $row ) {
 115                                 if ( !( ++$i % $reportingInterval ) ) {
 116                                         $this->output( "$i\n" );
 117                                         wfWaitForSlaves( $maxLag );
 118                                 }
 119                                 if($redirectsOnly)
 120                                         $this->fixRedirect( $row->page_id );
 121                                 else
 122                                         $this->fixLinksFromArticle( $row->page_id );
 123                         }
 124                 } else {
 125                         if ( !$end ) {
 126                                 $maxPage = $dbr->selectField( 'page', 'max(page_id)', false );
 127                                 $maxRD = $dbr->selectField( 'redirect', 'max(rd_from)', false );
 128                                 $end = max( $maxPage, $maxRD );
 129                         }
 130                         $this->output( "Refreshing redirects table.\n" );
 131                         $this->output( "Starting from page_id $start of $end.\n" );
 132
 133                         for ($id = $start; $id <= $end; $id++) {
 134
 135                                 if ( !($id % $reportingInterval) ) {
 136                                         $this->output( "$id\n" );
 137                                         wfWaitForSlaves( $maxLag );
 138                                 }
 139                                 $this->fixRedirect( $id );
 140                         }
 141
 142                         if(!$redirectsOnly) {
 143                                 $this->output( "Refreshing links table.\n" );
 144                                 $this->output( "Starting from page_id $start of $end.\n" );
 145
 146                                 for ($id = $start; $id <= $end; $id++) {
 147
 148                                         if ( !($id % $reportingInterval) ) {
 149                                                 $this->output( "$id\n" );
 150                                                 wfWaitForSlaves( $maxLag );
 151                                         }
 152                                         $this->fixLinksFromArticle( $id );
 153                                 }
 154                         }
 155                 }
 156         }
 157
 158         /**
 159          * Update the redirect entry for a given page
 160          * @param $id int The page_id of the redirect
 161          */
 162         private function fixRedirect( $id ){
 163                 global $wgTitle, $wgArticle;
 164
 165                 $wgTitle = Title::newFromID( $id );
 166                 $dbw = wfGetDB( DB_MASTER );
 167
 168                 if ( is_null( $wgTitle ) ) {
 169                         // This page doesn't exist (any more)
 170                         // Delete any redirect table entry for it
 171                         $dbw->delete( 'redirect', array( 'rd_from' => $id ),
 172                                 __METHOD__ );
 173                         return;
 174                 }
 175                 $wgArticle = new Article($wgTitle);
 176
 177                 $rt = $wgArticle->followRedirect();
 178
 179                 if($rt == false || !is_object($rt)) {
 180                         // $wgTitle is not a redirect
 181                         // Delete any redirect table entry for it
 182                         $dbw->delete( 'redirect', array( 'rd_from' => $id ),
 183                                 __METHOD__ );
 184                 } else {
 185                         $wgArticle->updateRedirectOn($dbw,$rt);
 186                 }
 187         }
 188
 189         /**
 190          * Run LinksUpdate for all links on a given page_id
 191          * @param $id int The page_id
 192          */
 193         private function fixLinksFromArticle( $id ) {
 194                 global $wgTitle, $wgParser;
 195
 196                 $wgTitle = Title::newFromID( $id );
 197                 $dbw = wfGetDB( DB_MASTER );
 198
 199                 $linkCache =& LinkCache::singleton();
 200                 $linkCache->clear();
 201
 202                 if ( is_null( $wgTitle ) ) {
 203                         return;
 204                 }
 205                 $dbw->begin();
 206
 207                 $revision = Revision::newFromTitle( $wgTitle );
 208                 if ( !$revision ) {
 209                         return;
 210                 }
 211
 212                 $options = new ParserOptions;
 213                 $parserOutput = $wgParser->parse( $revision->getText(), $wgTitle, $options, true, true, $revision->getId() );
 214                 $update = new LinksUpdate( $wgTitle, $parserOutput, false );
 215                 $update->doUpdate();
 216                 $dbw->commit();
 217         }
 218
 219         /*
 220          * Removes non-existing links from pages from pagelinks, imagelinks,
 221          * categorylinks, templatelinks and externallinks tables.
 222          *
 223          * @param $maxLag
 224          * @param $batchSize The size of deletion batches
 225          *
 226          * @author Merlijn van Deen <valhallasw@arctus.nl>
 227          */
 228         private function deleteLinksFromNonexistent( $maxLag = 0, $batchSize = 100 ) {
 229                 wfWaitForSlaves( $maxLag );
 230
 231                 $dbw = wfGetDB( DB_MASTER );
 232
 233                 $lb = wfGetLBFactory()->newMainLB();
 234                 $dbr = $lb->getConnection( DB_SLAVE );
 235                 $dbr->bufferResults( false );
 236
 237                 $linksTables = array( // table name => page_id field
 238                         'pagelinks' => 'pl_from',
 239                         'imagelinks' => 'il_from',
 240                         'categorylinks' => 'cl_from',
 241                         'templatelinks' => 'tl_from',
 242                         'externallinks' => 'el_from',
 243                 );
 244
 245                 foreach ( $linksTables as $table => $field ) {
 246                         $this->output( "Retrieving illegal entries from $table... " );
 247
 248                         // SELECT DISTINCT( $field ) FROM $table LEFT JOIN page ON $field=page_id WHERE page_id IS NULL;
 249                         $results = $dbr->select( array( $table, 'page' ),
 250                                                   $field,
 251                                                   array('page_id' => null ),
 252                                                   __METHOD__,
 253                                                   'DISTINCT',
 254                                                   array( 'page' => array( 'LEFT JOIN', "$field=page_id"))
 255                         );
 256
 257                         $counter = 0;
 258                         $list = array();
 259                         $this->output( "0.." );
 260
 261                         foreach( $results as $row ) {
 262                                 $counter++;
 263                                 $list[] = $row->$field;
 264                                 if ( ( $counter % $batchSize ) == 0 ) {
 265                                         wfWaitForSlaves(5);
 266                                         $dbw->delete( $table, array( $field => $list ), __METHOD__ );
 267
 268                                         $this->output( $counter . ".." );
 269                                         $list = array();
 270                                 }
 271                         }
 272                         $this->output( $counter );
 273                         if (count($list) > 0) {
 274                                 $dbw->delete( $table, array( $field => $list ), __METHOD__ );
 275                         }
 276                         $this->output( "\n" );
 277                 }
 278                 $lb->closeAll();
 279         }
 280 }
 281
 282 $maintClass = 'RefreshLinks';
 283 require_once( DO_MAINTENANCE );