includes/historyblob/ConcatenatedGzipHistoryBlob.php

   1 <?php
   2 /**
   3  * Efficient concatenated text storage.
   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  * Concatenated gzip (CGZ) storage
  25  * Improves compression ratio by concatenating like objects before gzipping
  26  */
  27 class ConcatenatedGzipHistoryBlob implements HistoryBlob {
  28         public $mVersion = 0;
  29         public $mCompressed = false;
  30         /**
  31          * @var array|string
  32          * @fixme Why are some methods treating it as an array, and others as a string, unconditionally?
  33          */
  34         public $mItems = [];
  35         public $mDefaultHash = '';
  36         public $mSize = 0;
  37         public $mMaxSize = 10000000;
  38         public $mMaxCount = 100;
  39
  40         public function __construct() {
  41                 if ( !function_exists( 'gzdeflate' ) ) {
  42                         throw new MWException( "Need zlib support to read or write this "
  43                                 . "kind of history object (ConcatenatedGzipHistoryBlob)\n" );
  44                 }
  45         }
  46
  47         /**
  48          * @param string $text
  49          * @return string
  50          */
  51         public function addItem( $text ) {
  52                 $this->uncompress();
  53                 $hash = md5( $text );
  54                 if ( !isset( $this->mItems[$hash] ) ) {
  55                         $this->mItems[$hash] = $text;
  56                         $this->mSize += strlen( $text );
  57                 }
  58                 return $hash;
  59         }
  60
  61         /**
  62          * @param string $hash
  63          * @return array|bool
  64          */
  65         public function getItem( $hash ) {
  66                 $this->uncompress();
  67                 if ( array_key_exists( $hash, $this->mItems ) ) {
  68                         return $this->mItems[$hash];
  69                 } else {
  70                         return false;
  71                 }
  72         }
  73
  74         /**
  75          * @param string $text
  76          * @return void
  77          */
  78         public function setText( $text ) {
  79                 $this->uncompress();
  80                 $this->mDefaultHash = $this->addItem( $text );
  81         }
  82
  83         /**
  84          * @return array|bool
  85          */
  86         public function getText() {
  87                 $this->uncompress();
  88                 return $this->getItem( $this->mDefaultHash );
  89         }
  90
  91         /**
  92          * Remove an item
  93          *
  94          * @param string $hash
  95          */
  96         public function removeItem( $hash ) {
  97                 $this->mSize -= strlen( $this->mItems[$hash] );
  98                 unset( $this->mItems[$hash] );
  99         }
 100
 101         /**
 102          * Compress the bulk data in the object
 103          */
 104         public function compress() {
 105                 if ( !$this->mCompressed ) {
 106                         $this->mItems = gzdeflate( serialize( $this->mItems ) );
 107                         $this->mCompressed = true;
 108                 }
 109         }
 110
 111         /**
 112          * Uncompress bulk data
 113          */
 114         public function uncompress() {
 115                 if ( $this->mCompressed ) {
 116                         $this->mItems = unserialize( gzinflate( $this->mItems ) );
 117                         $this->mCompressed = false;
 118                 }
 119         }
 120
 121         /**
 122          * @return array
 123          */
 124         function __sleep() {
 125                 $this->compress();
 126                 return [ 'mVersion', 'mCompressed', 'mItems', 'mDefaultHash' ];
 127         }
 128
 129         function __wakeup() {
 130                 $this->uncompress();
 131         }
 132
 133         /**
 134          * Helper function for compression jobs
 135          * Returns true until the object is "full" and ready to be committed
 136          *
 137          * @return bool
 138          */
 139         public function isHappy() {
 140                 return $this->mSize < $this->mMaxSize
 141                         && count( $this->mItems ) < $this->mMaxCount;
 142         }
 143 }
 144
 145 // phpcs:ignore Generic.CodeAnalysis.UnconditionalIfStatement.Found
 146 if ( false ) {
 147         // Blobs generated by MediaWiki < 1.5 on PHP 4 were serialized with the
 148         // class name coerced to lowercase. We can improve efficiency by adding
 149         // autoload entries for the lowercase variants of these classes (T166759).
 150         // The code below is never executed, but it is picked up by the AutoloadGenerator
 151         // parser, which scans for class_alias() calls.
 152         class_alias( ConcatenatedGzipHistoryBlob::class, 'concatenatedgziphistoryblob' );
 153 }