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                 $files = RepoGroup::singleton()->findFiles( array_keys( $images ) );
  86
  87                 $fit = true;
  88                 $count = 0;
  89                 $titles = array();
  90
  91                 $sha1s = array();
  92                 foreach ( $files as $file ) {
  93                         $sha1s[$file->getName()] = $file->getSha1();
  94                 }
  95
  96                 // find all files with the hashes, result format is: array( hash => array( dup1, dup2 ), hash1 => ... )
  97                 $filesBySha1s = RepoGroup::singleton()->findBySha1s( array_unique( array_values( $sha1s ) ) );
  98
  99                 // iterate over $images to handle continue param correct
 100                 foreach( $images as $image => $pageId ) {
 101                         if( !isset( $sha1s[$image] ) ) {
 102                                 continue; //file does not exist
 103                         }
 104                         $sha1 = $sha1s[$image];
 105                         $dupFiles = $filesBySha1s[$sha1];
 106                         if( $params['dir'] == 'descending' ) {
 107                                 $dupFiles = array_reverse( $dupFiles );
 108                         }
 109                         foreach ( $dupFiles as $dupFile ) {
 110                                 $dupName = $dupFile->getName();
 111                                 if( $image == $dupName ) {
 112                                         continue; //ignore the file itself
 113                                 }
 114                                 if( $skipUntilThisDup !== false && $dupName < $skipUntilThisDup ) {
 115                                         continue; //skip to pos after the image from continue param
 116                                 }
 117                                 $skipUntilThisDup = false;
 118                                 if ( ++$count > $params['limit'] ) {
 119                                         $fit = false; //break outer loop
 120                                         // We're one over limit which shows that
 121                                         // there are additional images to be had. Stop here...
 122                                         $this->setContinueEnumParameter( 'continue', $image . '|' . $dupName );
 123                                         break;
 124                                 }
 125                                 if ( !is_null( $resultPageSet ) ) {
 126                                         $titles[] = $file->getTitle();
 127                                 } else {
 128                                         $r = array(
 129                                                 'name' => $dupName,
 130                                                 'user' => $dupFile->getUser( 'text' ),
 131                                                 'timestamp' => wfTimestamp( TS_ISO_8601, $dupFile->getTimestamp() )
 132                                         );
 133                                         $fit = $this->addPageSubItem( $pageId, $r );
 134                                         if ( !$fit ) {
 135                                                 $this->setContinueEnumParameter( 'continue', $image . '|' . $dupName );
 136                                                 break;
 137                                         }
 138                                 }
 139                         }
 140                         if( !$fit ) {
 141                                 break;
 142                         }
 143                 }
 144                 if ( !is_null( $resultPageSet ) ) {
 145                         $resultPageSet->populateFromTitles( $titles );
 146                 }
 147         }
 148
 149         public function getAllowedParams() {
 150                 return array(
 151                         'limit' => array(
 152                                 ApiBase::PARAM_DFLT => 10,
 153                                 ApiBase::PARAM_TYPE => 'limit',
 154                                 ApiBase::PARAM_MIN => 1,
 155                                 ApiBase::PARAM_MAX => ApiBase::LIMIT_BIG1,
 156                                 ApiBase::PARAM_MAX2 => ApiBase::LIMIT_BIG2
 157                         ),
 158                         'continue' => null,
 159                         'dir' => array(
 160                                 ApiBase::PARAM_DFLT => 'ascending',
 161                                 ApiBase::PARAM_TYPE => array(
 162                                         'ascending',
 163                                         'descending'
 164                                 )
 165                         ),
 166                 );
 167         }
 168
 169         public function getParamDescription() {
 170                 return array(
 171                         'limit' => 'How many duplicate files to return',
 172                         'continue' => 'When more results are available, use this to continue',
 173                         'dir' => 'The direction in which to list',
 174                 );
 175         }
 176
 177         public function getResultProperties() {
 178                 return array(
 179                         '' => array(
 180                                 'name' => 'string',
 181                                 'user' => 'string',
 182                                 'timestamp' => 'timestamp'
 183                         )
 184                 );
 185         }
 186
 187         public function getDescription() {
 188                 return 'List all files that are duplicates of the given file(s) based on hash values';
 189         }
 190
 191         public function getPossibleErrors() {
 192                 return array_merge( parent::getPossibleErrors(), array(
 193                         array( 'code' => '_badcontinue', 'info' => 'Invalid continue param. You should pass the original value returned by the previous query' ),
 194                 ) );
 195         }
 196
 197         public function getExamples() {
 198                 return array(
 199                         'api.php?action=query&titles=File:Albert_Einstein_Head.jpg&prop=duplicatefiles',
 200                         'api.php?action=query&generator=allimages&prop=duplicatefiles',
 201                 );
 202         }
 203
 204         public function getHelpUrls() {
 205                 return 'https://www.mediawiki.org/wiki/API:Properties#duplicatefiles_.2F_df';
 206         }
 207
 208         public function getVersion() {
 209                 return __CLASS__ . ': $Id$';
 210         }
 211 }