includes/api/ApiQueryDuplicateFiles.php

   1 <?php
   2 /**
   3  *
   4  *
   5  * Created on Sep 27, 2008
   6  *
   7  * Copyright © 2008 Roan Kattouw "<Firstname>.<Lastname>@gmail.com"
   8  *
   9  * This program is free software; you can redistribute it and/or modify
  10  * it under the terms of the GNU General Public License as published by
  11  * the Free Software Foundation; either version 2 of the License, or
  12  * (at your option) any later version.
  13  *
  14  * This program is distributed in the hope that it will be useful,
  15  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  16  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  17  * GNU General Public License for more details.
  18  *
  19  * You should have received a copy of the GNU General Public License along
  20  * with this program; if not, write to the Free Software Foundation, Inc.,
  21  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  22  * http://www.gnu.org/copyleft/gpl.html
  23  *
  24  * @file
  25  */
  26
  27 /**
  28  * A query module to list duplicates of the given file(s)
  29  *
  30  * @ingroup API
  31  */
  32 class ApiQueryDuplicateFiles extends ApiQueryGeneratorBase {
  33
  34         public function __construct( $query, $moduleName ) {
  35                 parent::__construct( $query, $moduleName, 'df' );
  36         }
  37
  38         public function execute() {
  39                 $this->run();
  40         }
  41
  42         public function getCacheMode( $params ) {
  43                 return 'public';
  44         }
  45
  46         public function executeGenerator( $resultPageSet ) {
  47                 $this->run( $resultPageSet );
  48         }
  49
  50         /**
  51          * @param $resultPageSet ApiPageSet
  52          * @return
  53          */
  54         private function run( $resultPageSet = null ) {
  55                 $params = $this->extractRequestParams();
  56                 $namespaces = $this->getPageSet()->getAllTitlesByNamespace();
  57                 if ( empty( $namespaces[NS_FILE] ) ) {
  58                         return;
  59                 }
  60                 $images = $namespaces[NS_FILE];
  61
  62                 if( $params['dir'] == 'descending' ) {
  63                         $images = array_reverse( $images );
  64                 }
  65
  66                 $skipUntilThisDup = false;
  67                 if ( isset( $params['continue'] ) ) {
  68                         $cont = explode( '|', $params['continue'] );
  69                         if ( count( $cont ) != 2 ) {
  70                                 $this->dieUsage( 'Invalid continue param. You should pass the ' .
  71                                         'original value returned by the previous query', '_badcontinue' );
  72                         }
  73                         $fromImage = $cont[0];
  74                         $skipUntilThisDup = $cont[1];
  75                         // Filter out any images before $fromImage
  76                         foreach ( $images as $image => $pageId ) {
  77                                 if ( $image < $fromImage ) {
  78                                         unset( $images[$image] );
  79                                 } else {
  80                                         break;
  81                                 }
  82                         }
  83                 }
  84
  85                 $filesToFind = array_keys( $images );
  86                 if( $params['localonly'] ) {
  87                         $files = RepoGroup::singleton()->getLocalRepo()->findFiles( $filesToFind );
  88                 } else {
  89                         $files = RepoGroup::singleton()->findFiles( $filesToFind );
  90                 }
  91
  92                 $fit = true;
  93                 $count = 0;
  94                 $titles = array();
  95
  96                 $sha1s = array();
  97                 foreach ( $files as $file ) {
  98                         $sha1s[$file->getName()] = $file->getSha1();
  99                 }
 100
 101                 // find all files with the hashes, result format is: array( hash => array( dup1, dup2 ), hash1 => ... )
 102                 $filesToFindBySha1s = array_unique( array_values( $sha1s ) );
 103                 if( $params['localonly'] ) {
 104                         $filesBySha1s = RepoGroup::singleton()->getLocalRepo()->findBySha1s( $filesToFindBySha1s );
 105                 } else {
 106                         $filesBySha1s = RepoGroup::singleton()->findBySha1s( $filesToFindBySha1s );
 107                 }
 108
 109                 // iterate over $images to handle continue param correct
 110                 foreach( $images as $image => $pageId ) {
 111                         if( !isset( $sha1s[$image] ) ) {
 112                                 continue; //file does not exist
 113                         }
 114                         $sha1 = $sha1s[$image];
 115                         $dupFiles = $filesBySha1s[$sha1];
 116                         if( $params['dir'] == 'descending' ) {
 117                                 $dupFiles = array_reverse( $dupFiles );
 118                         }
 119                         foreach ( $dupFiles as $dupFile ) {
 120                                 $dupName = $dupFile->getName();
 121                                 if( $image == $dupName ) {
 122                                         continue; //ignore the file itself
 123                                 }
 124                                 if( $skipUntilThisDup !== false && $dupName < $skipUntilThisDup ) {
 125                                         continue; //skip to pos after the image from continue param
 126                                 }
 127                                 $skipUntilThisDup = false;
 128                                 if ( ++$count > $params['limit'] ) {
 129                                         $fit = false; //break outer loop
 130                                         // We're one over limit which shows that
 131                                         // there are additional images to be had. Stop here...
 132                                         $this->setContinueEnumParameter( 'continue', $image . '|' . $dupName );
 133                                         break;
 134                                 }
 135                                 if ( !is_null( $resultPageSet ) ) {
 136                                         $titles[] = $file->getTitle();
 137                                 } else {
 138                                         $r = array(
 139                                                 'name' => $dupName,
 140                                                 'user' => $dupFile->getUser( 'text' ),
 141                                                 'timestamp' => wfTimestamp( TS_ISO_8601, $dupFile->getTimestamp() )
 142                                         );
 143                                         $fit = $this->addPageSubItem( $pageId, $r );
 144                                         if ( !$fit ) {
 145                                                 $this->setContinueEnumParameter( 'continue', $image . '|' . $dupName );
 146                                                 break;
 147                                         }
 148                                 }
 149                         }
 150                         if( !$fit ) {
 151                                 break;
 152                         }
 153                 }
 154                 if ( !is_null( $resultPageSet ) ) {
 155                         $resultPageSet->populateFromTitles( $titles );
 156                 }
 157         }
 158
 159         public function getAllowedParams() {
 160                 return array(
 161                         'limit' => array(
 162                                 ApiBase::PARAM_DFLT => 10,
 163                                 ApiBase::PARAM_TYPE => 'limit',
 164                                 ApiBase::PARAM_MIN => 1,
 165                                 ApiBase::PARAM_MAX => ApiBase::LIMIT_BIG1,
 166                                 ApiBase::PARAM_MAX2 => ApiBase::LIMIT_BIG2
 167                         ),
 168                         'continue' => null,
 169                         'dir' => array(
 170                                 ApiBase::PARAM_DFLT => 'ascending',
 171                                 ApiBase::PARAM_TYPE => array(
 172                                         'ascending',
 173                                         'descending'
 174                                 )
 175                         ),
 176                         'localonly' => false,
 177                 );
 178         }
 179
 180         public function getParamDescription() {
 181                 return array(
 182                         'limit' => 'How many duplicate files to return',
 183                         'continue' => 'When more results are available, use this to continue',
 184                         'dir' => 'The direction in which to list',
 185                         'localonly' => 'Look only for files in the local repository',
 186                 );
 187         }
 188
 189         public function getResultProperties() {
 190                 return array(
 191                         '' => array(
 192                                 'name' => 'string',
 193                                 'user' => 'string',
 194                                 'timestamp' => 'timestamp'
 195                         )
 196                 );
 197         }
 198
 199         public function getDescription() {
 200                 return 'List all files that are duplicates of the given file(s) based on hash values';
 201         }
 202
 203         public function getPossibleErrors() {
 204                 return array_merge( parent::getPossibleErrors(), array(
 205                         array( 'code' => '_badcontinue', 'info' => 'Invalid continue param. You should pass the original value returned by the previous query' ),
 206                 ) );
 207         }
 208
 209         public function getExamples() {
 210                 return array(
 211                         'api.php?action=query&titles=File:Albert_Einstein_Head.jpg&prop=duplicatefiles',
 212                         'api.php?action=query&generator=allimages&prop=duplicatefiles',
 213                 );
 214         }
 215
 216         public function getHelpUrls() {
 217                 return 'https://www.mediawiki.org/wiki/API:Properties#duplicatefiles_.2F_df';
 218         }
 219
 220         public function getVersion() {
 221                 return __CLASS__ . ': $Id$';
 222         }
 223 }