maintenance/rebuildrecentchanges.php

   1 <?php
   2 /**
   3  * Rebuild link tracking tables from scratch.  This takes several
   4  * hours, depending on the database size and server configuration.
   5  *
   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.
  10  *
  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.
  15  *
  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
  20  *
  21  * @ingroup Maintenance
  22  * @todo Document
  23  */
  24
  25 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  26
  27 class RebuildRecentchanges extends Maintenance {
  28         public function __construct() {
  29                 parent::__construct();
  30                 $this->mDescription = "Rebuild recent changes";
  31         }
  32
  33         public function execute() {
  34                 $this->rebuildRecentChangesTablePass1();
  35                 $this->rebuildRecentChangesTablePass2();
  36                 $this->rebuildRecentChangesTablePass3();
  37                 $this->rebuildRecentChangesTablePass4();
  38                 $this->purgeFeeds();
  39                 $this->output( "Done.\n" );
  40         }
  41
  42         /**
  43          * Rebuild pass 1
  44          * DOCUMENT ME!
  45          */
  46         private function rebuildRecentChangesTablePass1() {
  47                 $dbw = wfGetDB( DB_MASTER );
  48
  49                 $dbw->delete( 'recentchanges', '*' );
  50
  51                 $this->output( "Loading from page and revision tables...\n" );
  52
  53                 global $wgRCMaxAge;
  54
  55                 $this->output( '$wgRCMaxAge=' . $wgRCMaxAge );
  56                 $days = $wgRCMaxAge / 24 / 3600;
  57                 if ( intval( $days ) == $days ) {
  58                                 $this->output( " (" . $days . " days)\n" );
  59                 } else {
  60                                 $this->output( " (approx. " .  intval( $days ) . " days)\n" );
  61                 }
  62
  63                 $cutoff = time() - $wgRCMaxAge;
  64                 $dbw->insertSelect( 'recentchanges', array( 'page', 'revision' ),
  65                         array(
  66                                 'rc_timestamp'  => 'rev_timestamp',
  67                                 'rc_cur_time'   => 'rev_timestamp',
  68                                 'rc_user'       => 'rev_user',
  69                                 'rc_user_text'  => 'rev_user_text',
  70                                 'rc_namespace'  => 'page_namespace',
  71                                 'rc_title'      => 'page_title',
  72                                 'rc_comment'    => 'rev_comment',
  73                                 'rc_minor'      => 'rev_minor_edit',
  74                                 'rc_bot'        => 0,
  75                                 'rc_new'        => 'page_is_new',
  76                                 'rc_cur_id'     => 'page_id',
  77                                 'rc_this_oldid' => 'rev_id',
  78                                 'rc_last_oldid' => 0, // is this ok?
  79                                 'rc_type'       => $dbw->conditional( 'page_is_new != 0', RC_NEW, RC_EDIT ),
  80                                 'rc_deleted'    => 'rev_deleted'
  81                         ), array(
  82                                 'rev_timestamp > ' . $dbw->addQuotes( $dbw->timestamp( $cutoff ) ),
  83                                 'rev_page=page_id'
  84                         ), __METHOD__,
  85                         array(), // INSERT options
  86                         array( 'ORDER BY' => 'rev_timestamp DESC', 'LIMIT' => 5000 ) // SELECT options
  87                 );
  88         }
  89
  90         /**
  91          * Rebuild pass 2
  92          * DOCUMENT ME!
  93          */
  94         private function rebuildRecentChangesTablePass2() {
  95                 $dbw = wfGetDB( DB_MASTER );
  96                 list ( $recentchanges, $revision ) = $dbw->tableNamesN( 'recentchanges', 'revision' );
  97
  98                 $this->output( "Updating links and size differences...\n" );
  99
 100                 # Fill in the rc_last_oldid field, which points to the previous edit
 101                 $sql = "SELECT rc_cur_id,rc_this_oldid,rc_timestamp FROM $recentchanges " .
 102                   "ORDER BY rc_cur_id,rc_timestamp";
 103                 $res = $dbw->query( $sql, DB_MASTER );
 104
 105                 $lastCurId = 0;
 106                 $lastOldId = 0;
 107                 foreach ( $res as $obj ) {
 108                         $new = 0;
 109                         if ( $obj->rc_cur_id != $lastCurId ) {
 110                                 # Switch! Look up the previous last edit, if any
 111                                 $lastCurId = intval( $obj->rc_cur_id );
 112                                 $emit = $obj->rc_timestamp;
 113                                 $sql2 = "SELECT rev_id,rev_len FROM $revision " .
 114                                         "WHERE rev_page={$lastCurId} " .
 115                                         "AND rev_timestamp<'{$emit}' ORDER BY rev_timestamp DESC";
 116                                 $sql2 = $dbw->limitResult( $sql2, 1, false );
 117                                 $res2 = $dbw->query( $sql2 );
 118                                 $row = $dbw->fetchObject( $res2 );
 119                                 if ( $row ) {
 120                                         $lastOldId = intval( $row->rev_id );
 121                                         # Grab the last text size if available
 122                                         $lastSize = !is_null( $row->rev_len ) ? intval( $row->rev_len ) : null;
 123                                 } else {
 124                                         # No previous edit
 125                                         $lastOldId = 0;
 126                                         $lastSize = null;
 127                                         $new = 1; // probably true
 128                                 }
 129                         }
 130                         if ( $lastCurId == 0 ) {
 131                                 $this->output( "Uhhh, something wrong? No curid\n" );
 132                         } else {
 133                                 # Grab the entry's text size
 134                                 $size = $dbw->selectField( 'revision', 'rev_len', array( 'rev_id' => $obj->rc_this_oldid ) );
 135
 136                                 $dbw->update( 'recentchanges',
 137                                         array(
 138                                                 'rc_last_oldid' => $lastOldId,
 139                                                 'rc_new'        => $new,
 140                                                 'rc_type'       => $new,
 141                                                 'rc_old_len'    => $lastSize,
 142                                                 'rc_new_len'    => $size,
 143                                         ), array(
 144                                                 'rc_cur_id'     => $lastCurId,
 145                                                 'rc_this_oldid' => $obj->rc_this_oldid,
 146                                         ),
 147                                         __METHOD__
 148                                 );
 149
 150                                 $lastOldId = intval( $obj->rc_this_oldid );
 151                                 $lastSize = $size;
 152                         }
 153                 }
 154         }
 155
 156         /**
 157          * Rebuild pass 3
 158          * DOCUMENT ME!
 159          */
 160         private function rebuildRecentChangesTablePass3() {
 161                 $dbw = wfGetDB( DB_MASTER );
 162
 163                 $this->output( "Loading from user, page, and logging tables...\n" );
 164
 165                 global $wgRCMaxAge, $wgLogTypes, $wgLogRestrictions;
 166                 // Some logs don't go in RC. This should check for that
 167                 $basicRCLogs = array_diff( $wgLogTypes, array_keys( $wgLogRestrictions ) );
 168
 169                 // Escape...blah blah
 170                 $selectLogs = array();
 171                 foreach ( $basicRCLogs as $logtype ) {
 172                         $safetype = $dbw->strencode( $logtype );
 173                         $selectLogs[] = "'$safetype'";
 174                 }
 175
 176                 $cutoff = time() - $wgRCMaxAge;
 177                 list( $logging, $page ) = $dbw->tableNamesN( 'logging', 'page' );
 178                 $dbw->insertSelect( 'recentchanges', array( 'user', "$logging LEFT JOIN $page ON (log_namespace=page_namespace AND log_title=page_title)" ),
 179                         array(
 180                                 'rc_timestamp'  => 'log_timestamp',
 181                                 'rc_cur_time'   => 'log_timestamp',
 182                                 'rc_user'       => 'log_user',
 183                                 'rc_user_text'  => 'user_name',
 184                                 'rc_namespace'  => 'log_namespace',
 185                                 'rc_title'      => 'log_title',
 186                                 'rc_comment'    => 'log_comment',
 187                                 'rc_minor'      => 0,
 188                                 'rc_bot'        => 0,
 189                                 'rc_patrolled'  => 1,
 190                                 'rc_new'        => 0,
 191                                 'rc_this_oldid' => 0,
 192                                 'rc_last_oldid' => 0,
 193                                 'rc_type'       => RC_LOG,
 194                                 'rc_cur_id'     => $dbw->cascadingDeletes() ? 'page_id' : 'COALESCE(page_id, 0)',
 195                                 'rc_log_type'   => 'log_type',
 196                                 'rc_log_action' => 'log_action',
 197                                 'rc_logid'      => 'log_id',
 198                                 'rc_params'     => 'log_params',
 199                                 'rc_deleted'    => 'log_deleted'
 200                         ), array(
 201                                 'log_timestamp > ' . $dbw->addQuotes( $dbw->timestamp( $cutoff ) ),
 202                                 'log_user=user_id',
 203                                 'log_type IN(' . implode( ',', $selectLogs ) . ')'
 204                         ), __METHOD__,
 205                         array(), // INSERT options
 206                         array( 'ORDER BY' => 'log_timestamp DESC', 'LIMIT' => 5000 ) // SELECT options
 207                 );
 208         }
 209
 210         /**
 211          * Rebuild pass 4
 212          * DOCUMENT ME!
 213          */
 214         private function rebuildRecentChangesTablePass4() {
 215                 global $wgGroupPermissions, $wgUseRCPatrol;
 216
 217                 $dbw = wfGetDB( DB_MASTER );
 218
 219                 list( $recentchanges, $usergroups, $user ) = $dbw->tableNamesN( 'recentchanges', 'user_groups', 'user' );
 220
 221                 $botgroups = $autopatrolgroups = array();
 222                 foreach ( $wgGroupPermissions as $group => $rights ) {
 223                         if ( isset( $rights['bot'] ) && $rights['bot'] ) {
 224                                 $botgroups[] = $dbw->addQuotes( $group );
 225                         }
 226                         if ( $wgUseRCPatrol && isset( $rights['autopatrol'] ) && $rights['autopatrol'] ) {
 227                                 $autopatrolgroups[] = $dbw->addQuotes( $group );
 228                         }
 229                 }
 230                 # Flag our recent bot edits
 231                 if ( !empty( $botgroups ) ) {
 232                         $botwhere = implode( ',', $botgroups );
 233                         $botusers = array();
 234
 235                         $this->output( "Flagging bot account edits...\n" );
 236
 237                         # Find all users that are bots
 238                         $sql = "SELECT DISTINCT user_name FROM $usergroups, $user " .
 239                                 "WHERE ug_group IN($botwhere) AND user_id = ug_user";
 240                         $res = $dbw->query( $sql, DB_MASTER );
 241
 242                         foreach ( $res as $obj ) {
 243                                 $botusers[] = $dbw->addQuotes( $obj->user_name );
 244                         }
 245                         # Fill in the rc_bot field
 246                         if ( !empty( $botusers ) ) {
 247                                 $botwhere = implode( ',', $botusers );
 248                                 $sql2 = "UPDATE $recentchanges SET rc_bot=1 " .
 249                                         "WHERE rc_user_text IN($botwhere)";
 250                                 $dbw->query( $sql2 );
 251                         }
 252                 }
 253                 global $wgMiserMode;
 254                 # Flag our recent autopatrolled edits
 255                 if ( !$wgMiserMode && !empty( $autopatrolgroups ) ) {
 256                         $patrolwhere = implode( ',', $autopatrolgroups );
 257                         $patrolusers = array();
 258
 259                         $this->output( "Flagging auto-patrolled edits...\n" );
 260
 261                         # Find all users in RC with autopatrol rights
 262                         $sql = "SELECT DISTINCT user_name FROM $usergroups, $user " .
 263                                 "WHERE ug_group IN($patrolwhere) AND user_id = ug_user";
 264                         $res = $dbw->query( $sql, DB_MASTER );
 265
 266                         foreach ( $res as $obj ) {
 267                                 $patrolusers[] = $dbw->addQuotes( $obj->user_name );
 268                         }
 269
 270                         # Fill in the rc_patrolled field
 271                         if ( !empty( $patrolusers ) ) {
 272                                 $patrolwhere = implode( ',', $patrolusers );
 273                                 $sql2 = "UPDATE $recentchanges SET rc_patrolled=1 " .
 274                                         "WHERE rc_user_text IN($patrolwhere)";
 275                                 $dbw->query( $sql2 );
 276                         }
 277                 }
 278         }
 279
 280         /**
 281          * Purge cached feeds in $messageMemc
 282          */
 283         private function purgeFeeds() {
 284                 global $wgFeedClasses, $messageMemc;
 285
 286                 $this->output( "Deleting feed timestamps.\n" );
 287
 288                 foreach ( $wgFeedClasses as $feed => $className ) {
 289                         $messageMemc->delete( wfMemcKey( 'rcfeed', $feed, 'timestamp' ) ); # Good enough for now.
 290                 }
 291         }
 292
 293 }
 294
 295 $maintClass = "RebuildRecentchanges";
 296 require_once( RUN_MAINTENANCE_IF_MAIN );