includes/api/ApiQueryDuplicateFiles.php

   1 <?php
   2 /**
   3  * Copyright © 2008 Roan Kattouw <roan.kattouw@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 namespace MediaWiki\Api;
  24
  25 use File;
  26 use RepoGroup;
  27 use Wikimedia\ParamValidator\ParamValidator;
  28 use Wikimedia\ParamValidator\TypeDef\IntegerDef;
  29
  30 /**
  31  * A query module to list duplicates of the given file(s)
  32  *
  33  * @ingroup API
  34  */
  35 class ApiQueryDuplicateFiles extends ApiQueryGeneratorBase {
  36
  37         private RepoGroup $repoGroup;
  38
  39         public function __construct(
  40                 ApiQuery $query,
  41                 string $moduleName,
  42                 RepoGroup $repoGroup
  43         ) {
  44                 parent::__construct( $query, $moduleName, 'df' );
  45                 $this->repoGroup = $repoGroup;
  46         }
  47
  48         public function execute() {
  49                 $this->run();
  50         }
  51
  52         public function getCacheMode( $params ) {
  53                 return 'public';
  54         }
  55
  56         public function executeGenerator( $resultPageSet ) {
  57                 $this->run( $resultPageSet );
  58         }
  59
  60         /**
  61          * @param ApiPageSet|null $resultPageSet
  62          */
  63         private function run( $resultPageSet = null ) {
  64                 $params = $this->extractRequestParams();
  65                 $namespaces = $this->getPageSet()->getGoodAndMissingTitlesByNamespace();
  66                 if ( empty( $namespaces[NS_FILE] ) ) {
  67                         return;
  68                 }
  69                 $images = $namespaces[NS_FILE];
  70
  71                 if ( $params['dir'] == 'descending' ) {
  72                         $images = array_reverse( $images );
  73                 }
  74
  75                 $skipUntilThisDup = false;
  76                 if ( isset( $params['continue'] ) ) {
  77                         $cont = $this->parseContinueParamOrDie( $params['continue'], [ 'string', 'string' ] );
  78                         $fromImage = $cont[0];
  79                         $skipUntilThisDup = $cont[1];
  80                         // Filter out any images before $fromImage
  81                         foreach ( $images as $image => $pageId ) {
  82                                 if ( $image < $fromImage ) {
  83                                         unset( $images[$image] );
  84                                 } else {
  85                                         break;
  86                                 }
  87                         }
  88                 }
  89
  90                 $filesToFind = array_keys( $images );
  91                 if ( $params['localonly'] ) {
  92                         $files = $this->repoGroup->getLocalRepo()->findFiles( $filesToFind );
  93                 } else {
  94                         $files = $this->repoGroup->findFiles( $filesToFind );
  95                 }
  96
  97                 $fit = true;
  98                 $count = 0;
  99                 $titles = [];
 100
 101                 $sha1s = [];
 102                 foreach ( $files as $file ) {
 103                         /** @var File $file */
 104                         $sha1s[$file->getName()] = $file->getSha1();
 105                 }
 106
 107                 // find all files with the hashes, result format is:
 108                 // [ hash => [ dup1, dup2 ], hash1 => ... ]
 109                 $filesToFindBySha1s = array_unique( array_values( $sha1s ) );
 110                 if ( $params['localonly'] ) {
 111                         $filesBySha1s = $this->repoGroup->getLocalRepo()->findBySha1s( $filesToFindBySha1s );
 112                 } else {
 113                         $filesBySha1s = $this->repoGroup->findBySha1s( $filesToFindBySha1s );
 114                 }
 115
 116                 // iterate over $images to handle continue param correct
 117                 foreach ( $images as $image => $pageId ) {
 118                         if ( !isset( $sha1s[$image] ) ) {
 119                                 continue; // file does not exist
 120                         }
 121                         $sha1 = $sha1s[$image];
 122                         $dupFiles = $filesBySha1s[$sha1];
 123                         if ( $params['dir'] == 'descending' ) {
 124                                 $dupFiles = array_reverse( $dupFiles );
 125                         }
 126                         /** @var File $dupFile */
 127                         foreach ( $dupFiles as $dupFile ) {
 128                                 $dupName = $dupFile->getName();
 129                                 if ( $image == $dupName && $dupFile->isLocal() ) {
 130                                         continue; // ignore the local file itself
 131                                 }
 132                                 if ( $skipUntilThisDup !== false && $dupName < $skipUntilThisDup ) {
 133                                         continue; // skip to pos after the image from continue param
 134                                 }
 135                                 $skipUntilThisDup = false;
 136                                 if ( ++$count > $params['limit'] ) {
 137                                         $fit = false; // break outer loop
 138                                         // We're one over limit which shows that
 139                                         // there are additional images to be had. Stop here...
 140                                         $this->setContinueEnumParameter( 'continue', $image . '|' . $dupName );
 141                                         break;
 142                                 }
 143                                 if ( $resultPageSet !== null ) {
 144                                         $titles[] = $dupFile->getTitle();
 145                                 } else {
 146                                         $r = [
 147                                                 'name' => $dupName,
 148                                                 'timestamp' => wfTimestamp( TS_ISO_8601, $dupFile->getTimestamp() ),
 149                                                 'shared' => !$dupFile->isLocal(),
 150                                         ];
 151                                         $uploader = $dupFile->getUploader( File::FOR_PUBLIC );
 152                                         if ( $uploader ) {
 153                                                 $r['user'] = $uploader->getName();
 154                                         }
 155                                         $fit = $this->addPageSubItem( $pageId, $r );
 156                                         if ( !$fit ) {
 157                                                 $this->setContinueEnumParameter( 'continue', $image . '|' . $dupName );
 158                                                 break;
 159                                         }
 160                                 }
 161                         }
 162                         if ( !$fit ) {
 163                                 break;
 164                         }
 165                 }
 166                 if ( $resultPageSet !== null ) {
 167                         $resultPageSet->populateFromTitles( $titles );
 168                 }
 169         }
 170
 171         public function getAllowedParams() {
 172                 return [
 173                         'limit' => [
 174                                 ParamValidator::PARAM_DEFAULT => 10,
 175                                 ParamValidator::PARAM_TYPE => 'limit',
 176                                 IntegerDef::PARAM_MIN => 1,
 177                                 IntegerDef::PARAM_MAX => ApiBase::LIMIT_BIG1,
 178                                 IntegerDef::PARAM_MAX2 => ApiBase::LIMIT_BIG2
 179                         ],
 180                         'continue' => [
 181                                 ApiBase::PARAM_HELP_MSG => 'api-help-param-continue',
 182                         ],
 183                         'dir' => [
 184                                 ParamValidator::PARAM_DEFAULT => 'ascending',
 185                                 ParamValidator::PARAM_TYPE => [
 186                                         'ascending',
 187                                         'descending'
 188                                 ]
 189                         ],
 190                         'localonly' => false,
 191                 ];
 192         }
 193
 194         protected function getExamplesMessages() {
 195                 return [
 196                         'action=query&titles=File:Albert_Einstein_Head.jpg&prop=duplicatefiles'
 197                                 => 'apihelp-query+duplicatefiles-example-simple',
 198                         'action=query&generator=allimages&prop=duplicatefiles'
 199                                 => 'apihelp-query+duplicatefiles-example-generated',
 200                 ];
 201         }
 202
 203         public function getHelpUrls() {
 204                 return 'https://www.mediawiki.org/wiki/Special:MyLanguage/API:Duplicatefiles';
 205         }
 206 }
 207
 208 /** @deprecated class alias since 1.43 */
 209 class_alias( ApiQueryDuplicateFiles::class, 'ApiQueryDuplicateFiles' );