maintenance/copyFileBackend.php

   1 <?php
   2 /**
   3  * Copy all files in some containers of one backend to another.
   4  *
   5  * This can also be used to re-shard the files for one backend using the
   6  * config of second backend. The second backend should have the same config
   7  * as the first, except for it having a different name and different sharding
   8  * configuration. The backend should be made read-only while this runs.
   9  * After this script finishes, the old files in the containers can be deleted.
  10  *
  11  * This program is free software; you can redistribute it and/or modify
  12  * it under the terms of the GNU General Public License as published by
  13  * the Free Software Foundation; either version 2 of the License, or
  14  * (at your option) any later version.
  15  *
  16  * This program is distributed in the hope that it will be useful,
  17  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  18  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  19  * GNU General Public License for more details.
  20  *
  21  * You should have received a copy of the GNU General Public License along
  22  * with this program; if not, write to the Free Software Foundation, Inc.,
  23  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  24  * http://www.gnu.org/copyleft/gpl.html
  25  *
  26  * @ingroup Maintenance
  27  */
  28
  29 require_once( dirname( __FILE__ ) . '/Maintenance.php' );
  30
  31 class CopyFileBackend extends Maintenance {
  32         public function __construct() {
  33                 parent::__construct();
  34                 $this->mDescription = "Copy files in one backend to another.";
  35                 $this->addOption( 'src', 'Backend containing the source files', true, true );
  36                 $this->addOption( 'dst', 'Backend where files should be copied to', true, true );
  37                 $this->addOption( 'containers', 'Pipe separated list of containers', true, true );
  38                 $this->addOption( 'subdir', 'Only do items in this child directory', false, true );
  39                 $this->setBatchSize( 50 );
  40         }
  41
  42         public function execute() {
  43                 $src = FileBackendGroup::singleton()->get( $this->getOption( 'src' ) );
  44                 $dst = FileBackendGroup::singleton()->get( $this->getOption( 'dst' ) );
  45                 $containers = explode( '|', $this->getOption( 'containers' ) );
  46                 $subDir = $this->getOption( rtrim( 'subdir', '/' ), '' );
  47
  48                 $count = 0;
  49                 foreach ( $containers as $container ) {
  50                         if ( $subDir != '' ) {
  51                                 $backendRel = "$container/$subDir";
  52                                 $this->output( "Doing container '$container', directory '$subDir'...\n" );
  53                         } else {
  54                                 $backendRel = $container;
  55                                 $this->output( "Doing container '$container'...\n" );
  56                         }
  57
  58                         $dir = $src->getRootStoragePath() . "/$backendRel";
  59                         $srcPathsRel = $src->getFileList( array( 'dir' => $dir ) );
  60                         if ( $srcPathsRel === null ) {
  61                                 $this->error( "Could not list files in $container.", 1 ); // die
  62                         }
  63
  64                         $batchPaths = array();
  65                         foreach ( $srcPathsRel as $srcPathRel ) {
  66                                 $batchPaths[$srcPathRel] = 1; // remove duplicates
  67                                 if ( count( $batchPaths ) >= $this->mBatchSize ) {
  68                                         $this->copyFileBatch( array_keys( $batchPaths ), $backendRel, $src, $dst );
  69                                         $batchPaths = array(); // done
  70                                 }
  71                                 ++$count;
  72                         }
  73                         if ( count( $batchPaths ) ) { // left-overs
  74                                 $this->copyFileBatch( array_keys( $batchPaths ), $backendRel, $src, $dst );
  75                                 $batchPaths = array(); // done
  76                         }
  77
  78                         if ( $subDir != '' ) {
  79                                 $this->output( "Finished container '$container', directory '$subDir'.\n" );
  80                         } else {
  81                                 $this->output( "Finished container '$container'.\n" );
  82                         }
  83                 }
  84
  85                 $this->output( "Done [$count file(s)].\n" );
  86         }
  87
  88         protected function copyFileBatch(
  89                 array $srcPathsRel, $backendRel, FileBackend $src, FileBackend $dst
  90         ) {
  91                 $ops = array();
  92                 $fsFiles = array();
  93                 foreach ( $srcPathsRel as $srcPathRel ) {
  94                         $srcPath = $src->getRootStoragePath() . "/$backendRel/$srcPathRel";
  95                         $dstPath = $dst->getRootStoragePath() . "/$backendRel/$srcPathRel";
  96                         if ( $this->filesAreSame( $src, $dst, $srcPath, $dstPath ) ) {
  97                                 $this->output( "Already have $srcPathRel.\n" );
  98                                 continue; // assume already copied...
  99                         }
 100                         // Note: getLocalReference() is fast for FS backends
 101                         $fsFile = $src->getLocalReference( array( 'src' => $srcPath, 'latest' => 1 ) );
 102                         if ( !$fsFile ) {
 103                                 $this->error( "Could not get local copy of $srcPath.", 1 ); // die
 104                         }
 105                         $fsFiles[] = $fsFile; // keep TempFSFile objects alive as needed
 106                         // Note: prepare() is usually fast for key/value backends
 107                         $status = $dst->prepare( array( 'dir' => dirname( $dstPath ) ) );
 108                         if ( !$status->isOK() ) {
 109                                 $this->error( print_r( $status->getErrorsArray(), true ) );
 110                                 $this->error( "Could not copy $srcPath to $dstPath.", 1 ); // die
 111                         }
 112                         $ops[] = array( 'op' => 'store',
 113                                 'src' => $fsFile->getPath(), 'dst' => $dstPath, 'overwrite' => 1 );
 114                 }
 115
 116                 $status = $dst->doOperations( $ops, array( 'nonJournaled' => 1 ) );
 117                 if ( !$status->isOK() ) {
 118                         $this->error( print_r( $status->getErrorsArray(), true ) );
 119                         $this->error( "Could not copy file batch.", 1 ); // die
 120                 } else {
 121                         $this->output( "Copied these file(s):\n" . implode( "\n", $srcPathsRel ) . "\n\n" );
 122                 }
 123         }
 124
 125         protected function filesAreSame( FileBackend $src, FileBackend $dst, $sPath, $dPath ) {
 126                 return (
 127                         ( $src->fileExists( array( 'src' => $sPath, 'latest' => 1 ) )
 128                                 === $dst->fileExists( array( 'src' => $dPath, 'latest' => 1 ) ) // short-circuit
 129                         ) && ( $src->getFileSize( array( 'src' => $sPath, 'latest' => 1 ) )
 130                                 === $dst->getFileSize( array( 'src' => $dPath, 'latest' => 1 ) ) // short-circuit
 131                         ) && ( $src->getFileSha1Base36( array( 'src' => $sPath, 'latest' => 1 ) )
 132                                 === $dst->getFileSha1Base36( array( 'src' => $dPath, 'latest' => 1 ) )
 133                         )
 134                 );
 135         }
 136 }
 137
 138 $maintClass = 'CopyFileBackend';
 139 require_once( RUN_MAINTENANCE_IF_MAIN );