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                 global $wgTitle;
  35                 $wgTitle = Title::newFromText( "Rebuild recent changes script" );
  36                 $this->rebuildRecentChangesTablePass1();
  37                 $this->rebuildRecentChangesTablePass2();
  38                 $this->rebuildRecentChangesTablePass3();
  39                 $this->rebuildRecentChangesTablePass4();
  40                 $this->purgeFeeds();
  41                 $this->output( "Done.\n" );
  42         }
  43
  44         /**
  45          * Rebuild pass 1
  46          * DOCUMENT ME!
  47          */
  48         function rebuildRecentChangesTablePass1()
  49         {
  50                 $dbw = wfGetDB( DB_MASTER );
  51
  52                 $dbw->delete( 'recentchanges', '*' );
  53
  54                 $this->output( "Loading from page and revision tables...\n" );
  55
  56                 global $wgRCMaxAge;
  57
  58                 $this->output( '$wgRCMaxAge=' . $wgRCMaxAge );
  59                 $days = $wgRCMaxAge / 24 / 3600;
  60                 if ( intval( $days ) == $days ) {
  61                                 $this->output( " (" . $days . " days)\n" );
  62                 } else {
  63                                 $this->output( " (approx. " .  intval( $days ) . " days)\n" );
  64                 }
  65
  66                 $cutoff = time() - $wgRCMaxAge;
  67                 $dbw->insertSelect( 'recentchanges', array( 'page', 'revision' ),
  68                         array(
  69                                 'rc_timestamp'  => 'rev_timestamp',
  70                                 'rc_cur_time'   => 'rev_timestamp',
  71                                 'rc_user'       => 'rev_user',
  72                                 'rc_user_text'  => 'rev_user_text',
  73                                 'rc_namespace'  => 'page_namespace',
  74                                 'rc_title'      => 'page_title',
  75                                 'rc_comment'    => 'rev_comment',
  76                                 'rc_minor'      => 'rev_minor_edit',
  77                                 'rc_bot'        => 0,
  78                                 'rc_new'        => 'page_is_new',
  79                                 'rc_cur_id'     => 'page_id',
  80                                 'rc_this_oldid' => 'rev_id',
  81                                 'rc_last_oldid' => 0, // is this ok?
  82                                 'rc_type'       => $dbw->conditional( 'page_is_new != 0', RC_NEW, RC_EDIT ),
  83                                 'rc_deleted'    => 'rev_deleted'
  84                         ), array(
  85                                 'rev_timestamp > ' . $dbw->addQuotes( $dbw->timestamp( $cutoff ) ),
  86                                 'rev_page=page_id'
  87                         ), __METHOD__,
  88                         array(), // INSERT options
  89                         array( 'ORDER BY' => 'rev_timestamp DESC', 'LIMIT' => 5000 ) // SELECT options
  90                 );
  91         }
  92
  93         /**
  94          * Rebuild pass 2
  95          * DOCUMENT ME!
  96          */
  97         private function rebuildRecentChangesTablePass2() {
  98                 $dbw = wfGetDB( DB_MASTER );
  99                 list ( $recentchanges, $revision ) = $dbw->tableNamesN( 'recentchanges', 'revision' );
 100
 101                 $this->output( "Updating links and size differences...\n" );
 102
 103                 # Fill in the rc_last_oldid field, which points to the previous edit
 104                 $sql = "SELECT rc_cur_id,rc_this_oldid,rc_timestamp FROM $recentchanges " .
 105                   "ORDER BY rc_cur_id,rc_timestamp";
 106                 $res = $dbw->query( $sql, DB_MASTER );
 107
 108                 $lastCurId = 0;
 109                 $lastOldId = 0;
 110                 foreach ( $res as $obj ) {
 111                         $new = 0;
 112                         if ( $obj->rc_cur_id != $lastCurId ) {
 113                                 # Switch! Look up the previous last edit, if any
 114                                 $lastCurId = intval( $obj->rc_cur_id );
 115                                 $emit = $obj->rc_timestamp;
 116                                 $sql2 = "SELECT rev_id,rev_len FROM $revision " .
 117                                         "WHERE rev_page={$lastCurId} " .
 118                                         "AND rev_timestamp<'{$emit}' ORDER BY rev_timestamp DESC";
 119                                 $sql2 = $dbw->limitResult( $sql2, 1, false );
 120                                 $res2 = $dbw->query( $sql2 );
 121                                 if ( $row = $dbw->fetchObject( $res2 ) ) {
 122                                         $lastOldId = intval( $row->rev_id );
 123                                         # Grab the last text size if available
 124                                         $lastSize = !is_null( $row->rev_len ) ? intval( $row->rev_len ) : 'NULL';
 125                                 } else {
 126                                         # No previous edit
 127                                         $lastOldId = 0;
 128                                         $lastSize = 'NULL';
 129                                         $new = 1; // probably true
 130                                 }
 131                                 $dbw->freeResult( $res2 );
 132                         }
 133                         if ( $lastCurId == 0 ) {
 134                                 $this->output( "Uhhh, something wrong? No curid\n" );
 135                         } else {
 136                                 # Grab the entry's text size
 137                                 $size = $dbw->selectField( 'revision', 'rev_len', array( 'rev_id' => $obj->rc_this_oldid ) );
 138                                 $size = !is_null( $size ) ? intval( $size ) : 'NULL';
 139
 140                                 $sql3 = "UPDATE $recentchanges SET rc_last_oldid=$lastOldId,rc_new=$new,rc_type=$new," .
 141                                         "rc_old_len=$lastSize,rc_new_len=$size " .
 142                                         "WHERE rc_cur_id={$lastCurId} AND rc_this_oldid={$obj->rc_this_oldid}";
 143                                 $dbw->query( $sql3 );
 144
 145                                 $lastOldId = intval( $obj->rc_this_oldid );
 146                                 $lastSize = $size;
 147                         }
 148                 }
 149                 $dbw->freeResult( $res );
 150         }
 151
 152         /**
 153          * Rebuild pass 3
 154          * DOCUMENT ME!
 155          */
 156         private function rebuildRecentChangesTablePass3() {
 157                 $dbw = wfGetDB( DB_MASTER );
 158
 159                 $this->output( "Loading from user, page, and logging tables...\n" );
 160
 161                 global $wgRCMaxAge, $wgLogTypes, $wgLogRestrictions;
 162                 // Some logs don't go in RC. This should check for that
 163                 $basicRCLogs = array_diff( $wgLogTypes, array_keys( $wgLogRestrictions ) );
 164
 165                 // Escape...blah blah
 166                 $selectLogs = array();
 167                 foreach ( $basicRCLogs as $logtype ) {
 168                         $safetype = $dbw->strencode( $logtype );
 169                         $selectLogs[] = "'$safetype'";
 170                 }
 171
 172                 $cutoff = time() - $wgRCMaxAge;
 173                 list( $logging, $page ) = $dbw->tableNamesN( 'logging', 'page' );
 174                 $dbw->insertSelect( 'recentchanges', array( 'user', "$logging LEFT JOIN $page ON (log_namespace=page_namespace AND log_title=page_title)" ),
 175                         array(
 176                                 'rc_timestamp'  => 'log_timestamp',
 177                                 'rc_cur_time'   => 'log_timestamp',
 178                                 'rc_user'       => 'log_user',
 179                                 'rc_user_text'  => 'user_name',
 180                                 'rc_namespace'  => 'log_namespace',
 181                                 'rc_title'      => 'log_title',
 182                                 'rc_comment'    => 'log_comment',
 183                                 'rc_minor'      => 0,
 184                                 'rc_bot'        => 0,
 185                                 'rc_patrolled'  => 1,
 186                                 'rc_new'        => 0,
 187                                 'rc_this_oldid' => 0,
 188                                 'rc_last_oldid' => 0,
 189                                 'rc_type'       => RC_LOG,
 190                                 'rc_cur_id'     => $dbw->cascadingDeletes() ? 'page_id' : 'COALESCE(page_id, 0)',
 191                                 'rc_log_type'   => 'log_type',
 192                                 'rc_log_action' => 'log_action',
 193                                 'rc_logid'      => 'log_id',
 194                                 'rc_params'     => 'log_params',
 195                                 'rc_deleted'    => 'log_deleted'
 196                         ), array(
 197                                 'log_timestamp > ' . $dbw->addQuotes( $dbw->timestamp( $cutoff ) ),
 198                                 'log_user=user_id',
 199                                 'log_type IN(' . implode( ',', $selectLogs ) . ')'
 200                         ), __METHOD__,
 201                         array(), // INSERT options
 202                         array( 'ORDER BY' => 'log_timestamp DESC', 'LIMIT' => 5000 ) // SELECT options
 203                 );
 204         }
 205
 206         /**
 207          * Rebuild pass 4
 208          * DOCUMENT ME!
 209          */
 210         private function rebuildRecentChangesTablePass4() {
 211                 global $wgGroupPermissions, $wgUseRCPatrol;
 212
 213                 $dbw = wfGetDB( DB_MASTER );
 214
 215                 list( $recentchanges, $usergroups, $user ) = $dbw->tableNamesN( 'recentchanges', 'user_groups', 'user' );
 216
 217                 $botgroups = $autopatrolgroups = array();
 218                 foreach ( $wgGroupPermissions as $group => $rights ) {
 219                         if ( isset( $rights['bot'] ) && $rights['bot'] ) {
 220                                 $botgroups[] = $dbw->addQuotes( $group );
 221                         }
 222                         if ( $wgUseRCPatrol && isset( $rights['autopatrol'] ) && $rights['autopatrol'] ) {
 223                                 $autopatrolgroups[] = $dbw->addQuotes( $group );
 224                         }
 225                 }
 226                 # Flag our recent bot edits
 227                 if ( !empty( $botgroups ) ) {
 228                         $botwhere = implode( ',', $botgroups );
 229                         $botusers = array();
 230
 231                         $this->output( "Flagging bot account edits...\n" );
 232
 233                         # Find all users that are bots
 234                         $sql = "SELECT DISTINCT user_name FROM $usergroups, $user " .
 235                                 "WHERE ug_group IN($botwhere) AND user_id = ug_user";
 236                         $res = $dbw->query( $sql, DB_MASTER );
 237
 238                         foreach ( $res as $obj ) {
 239                                 $botusers[] = $dbw->addQuotes( $obj->user_name );
 240                         }
 241                         # Fill in the rc_bot field
 242                         if ( !empty( $botusers ) ) {
 243                                 $botwhere = implode( ',', $botusers );
 244                                 $sql2 = "UPDATE $recentchanges SET rc_bot=1 " .
 245                                         "WHERE rc_user_text IN($botwhere)";
 246                                 $dbw->query( $sql2 );
 247                         }
 248                 }
 249                 global $wgMiserMode;
 250                 # Flag our recent autopatrolled edits
 251                 if ( !$wgMiserMode && !empty( $autopatrolgroups ) ) {
 252                         $patrolwhere = implode( ',', $autopatrolgroups );
 253                         $patrolusers = array();
 254
 255                         $this->output( "Flagging auto-patrolled edits...\n" );
 256
 257                         # Find all users in RC with autopatrol rights
 258                         $sql = "SELECT DISTINCT user_name FROM $usergroups, $user " .
 259                                 "WHERE ug_group IN($patrolwhere) AND user_id = ug_user";
 260                         $res = $dbw->query( $sql, DB_MASTER );
 261
 262                         foreach ( $res as $obj ) {
 263                                 $patrolusers[] = $dbw->addQuotes( $obj->user_name );
 264                         }
 265
 266                         # Fill in the rc_patrolled field
 267                         if ( !empty( $patrolusers ) ) {
 268                                 $patrolwhere = implode( ',', $patrolusers );
 269                                 $sql2 = "UPDATE $recentchanges SET rc_patrolled=1 " .
 270                                         "WHERE rc_user_text IN($patrolwhere)";
 271                                 $dbw->query( $sql2 );
 272                         }
 273                 }
 274
 275                 $dbw->freeResult( $res );
 276         }
 277
 278         /**
 279          * Purge cached feeds in $messageMemc
 280          */
 281         private function purgeFeeds() {
 282                 global $wgFeedClasses, $messageMemc;
 283
 284                 $this->output( "Deleting feed timestamps.\n" );
 285
 286                 foreach ( $wgFeedClasses as $feed => $className ) {
 287                         $messageMemc->delete( wfMemcKey( 'rcfeed', $feed, 'timestamp' ) ); # Good enough for now.
 288                 }
 289         }
 290
 291 }
 292
 293 $maintClass = "RebuildRecentchanges";
 294 require_once( DO_MAINTENANCE );