includes/utils/FileContentsHasher.php

   1 <?php
   2
   3 use Wikimedia\ObjectCache\APCUBagOStuff;
   4 use Wikimedia\ObjectCache\BagOStuff;
   5 use Wikimedia\ObjectCache\EmptyBagOStuff;
   6
   7 /**
   8  * Generate hash digests of file contents to help with cache invalidation.
   9  *
  10  * This program is free software; you can redistribute it and/or modify
  11  * it under the terms of the GNU General Public License as published by
  12  * the Free Software Foundation; either version 2 of the License, or
  13  * (at your option) any later version.
  14  *
  15  * This program is distributed in the hope that it will be useful,
  16  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  17  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  18  * GNU General Public License for more details.
  19  *
  20  * You should have received a copy of the GNU General Public License along
  21  * with this program; if not, write to the Free Software Foundation, Inc.,
  22  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  23  * http://www.gnu.org/copyleft/gpl.html
  24  *
  25  * @file
  26  */
  27 class FileContentsHasher {
  28         private const ALGO = 'md4';
  29
  30         /** @var BagOStuff */
  31         protected $cache;
  32
  33         /** @var FileContentsHasher */
  34         private static $instance;
  35
  36         public function __construct() {
  37                 $this->cache = function_exists( 'apcu_fetch' ) ? new APCUBagOStuff() : new EmptyBagOStuff();
  38         }
  39
  40         /**
  41          * Get the singleton instance of this class.
  42          *
  43          * @return FileContentsHasher
  44          */
  45         public static function singleton() {
  46                 if ( !self::$instance ) {
  47                         self::$instance = new self;
  48                 }
  49
  50                 return self::$instance;
  51         }
  52
  53         /**
  54          * Get a hash of a file's contents, either by retrieving a previously-
  55          * computed hash from the cache, or by computing a hash from the file.
  56          *
  57          * @param string $filePath Full path to the file.
  58          * @return string|bool Hash of file contents, or false if the file could not be read.
  59          */
  60         private function getFileContentsHashInternal( $filePath ) {
  61                 // phpcs:ignore Generic.PHP.NoSilencedErrors.Discouraged
  62                 $mtime = @filemtime( $filePath );
  63                 if ( $mtime === false ) {
  64                         return false;
  65                 }
  66
  67                 $cacheKey = $this->cache->makeGlobalKey( __CLASS__, $filePath, $mtime, self::ALGO );
  68                 return $this->cache->getWithSetCallback(
  69                         $cacheKey,
  70                         $this->cache::TTL_DAY,
  71                         static function () use ( $filePath ) {
  72                                 // phpcs:ignore Generic.PHP.NoSilencedErrors.Discouraged
  73                                 $contents = @file_get_contents( $filePath );
  74                                 if ( $contents === false ) {
  75                                         // Don't cache false
  76                                         return false;
  77                                 }
  78
  79                                 return hash( self::ALGO, $contents );
  80                         }
  81                 );
  82         }
  83
  84         /**
  85          * Get a hash of the combined contents of one or more files, either by
  86          * retrieving a previously-computed hash from the cache, or by computing
  87          * a hash from the files.
  88          *
  89          * @param string|string[] $filePaths One or more file paths.
  90          * @return string|bool Hash of files' contents, or false if no file could not be read.
  91          */
  92         public static function getFileContentsHash( $filePaths ) {
  93                 $instance = self::singleton();
  94
  95                 if ( !is_array( $filePaths ) ) {
  96                         $filePaths = (array)$filePaths;
  97                 }
  98
  99                 if ( count( $filePaths ) === 1 ) {
 100                         $hash = $instance->getFileContentsHashInternal( $filePaths[0] );
 101                         return $hash;
 102                 }
 103
 104                 sort( $filePaths );
 105                 $hashes = [];
 106                 foreach ( $filePaths as $filePath ) {
 107                         $hashes[] = $instance->getFileContentsHashInternal( $filePath ) ?: '';
 108                 }
 109
 110                 $hashes = implode( '', $hashes );
 111                 return $hashes ? hash( self::ALGO, $hashes ) : false;
 112         }
 113 }