includes/api/ApiQueryDuplicateFiles.php

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