includes/filerepo/backend/FileBackend.php

   1 <?php
   2 /**
   3  * @defgroup FileBackend File backend
   4  * @ingroup  FileRepo
   5  *
   6  * File backend is used to interact with file storage systems,
   7  * such as the local file system, NFS, or cloud storage systems.
   8  */
   9
  10 /**
  11  * Base class for all file backends.
  12  *
  13  * This program is free software; you can redistribute it and/or modify
  14  * it under the terms of the GNU General Public License as published by
  15  * the Free Software Foundation; either version 2 of the License, or
  16  * (at your option) any later version.
  17  *
  18  * This program is distributed in the hope that it will be useful,
  19  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  20  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  21  * GNU General Public License for more details.
  22  *
  23  * You should have received a copy of the GNU General Public License along
  24  * with this program; if not, write to the Free Software Foundation, Inc.,
  25  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  26  * http://www.gnu.org/copyleft/gpl.html
  27  *
  28  * @file
  29  * @ingroup FileBackend
  30  * @author Aaron Schulz
  31  */
  32
  33 /**
  34  * @brief Base class for all file backend classes (including multi-write backends).
  35  *
  36  * This class defines the methods as abstract that subclasses must implement.
  37  * Outside callers can assume that all backends will have these functions.
  38  *
  39  * All "storage paths" are of the format "mwstore://<backend>/<container>/<path>".
  40  * The "<path>" portion is a relative path that uses UNIX file system (FS)
  41  * notation, though any particular backend may not actually be using a local
  42  * filesystem.
  43  * Therefore, the relative paths are only virtual.
  44  *
  45  * Backend contents are stored under wiki-specific container names by default.
  46  * For legacy reasons, this has no effect for the FS backend class, and per-wiki
  47  * segregation must be done by setting the container paths appropriately.
  48  *
  49  * FS-based backends are somewhat more restrictive due to the existence of real
  50  * directory files; a regular file cannot have the same name as a directory. Other
  51  * backends with virtual directories may not have this limitation. Callers should
  52  * store files in such a way that no files and directories are under the same path.
  53  *
  54  * Methods should avoid throwing exceptions at all costs.
  55  * As a corollary, external dependencies should be kept to a minimum.
  56  *
  57  * @ingroup FileBackend
  58  * @since 1.19
  59  */
  60 abstract class FileBackend {
  61         protected $name; // string; unique backend name
  62         protected $wikiId; // string; unique wiki name
  63         protected $readOnly; // string; read-only explanation message
  64         protected $parallelize; // string; when to do operations in parallel
  65         protected $concurrency; // integer; how many operations can be done in parallel
  66
  67         /** @var LockManager */
  68         protected $lockManager;
  69         /** @var FileJournal */
  70         protected $fileJournal;
  71
  72         /**
  73          * Create a new backend instance from configuration.
  74          * This should only be called from within FileBackendGroup.
  75          *
  76          * $config includes:
  77          *   - name        : The unique name of this backend.
  78          *                   This should consist of alphanumberic, '-', and '_' characters.
  79          *                   This name should not be changed after use.
  80          *   - wikiId      : Prefix to container names that is unique to this wiki.
  81          *                   It should only consist of alphanumberic, '-', and '_' characters.
  82          *   - lockManager : Registered name of a file lock manager to use.
  83          *   - fileJournal : File journal configuration; see FileJournal::factory().
  84          *                   Journals simply log changes to files stored in the backend.
  85          *   - readOnly    : Write operations are disallowed if this is a non-empty string.
  86          *                   It should be an explanation for the backend being read-only.
  87          *   - parallelize : When to do file operations in parallel (when possible).
  88          *                   Allowed values are "implicit", "explicit" and "off".
  89          *   - concurrency : How many file operations can be done in parallel.
  90          *
  91          * @param $config Array
  92          * @throws MWException
  93          */
  94         public function __construct( array $config ) {
  95                 $this->name = $config['name'];
  96                 if ( !preg_match( '!^[a-zA-Z0-9-_]{1,255}$!', $this->name ) ) {
  97                         throw new MWException( "Backend name `{$this->name}` is invalid." );
  98                 }
  99                 $this->wikiId = isset( $config['wikiId'] )
 100                         ? $config['wikiId']
 101                         : wfWikiID(); // e.g. "my_wiki-en_"
 102                 $this->lockManager = ( $config['lockManager'] instanceof LockManager )
 103                         ? $config['lockManager']
 104                         : LockManagerGroup::singleton()->get( $config['lockManager'] );
 105                 $this->fileJournal = isset( $config['fileJournal'] )
 106                         ? ( ( $config['fileJournal'] instanceof FileJournal )
 107                                 ? $config['fileJournal']
 108                                 : FileJournal::factory( $config['fileJournal'], $this->name ) )
 109                         : FileJournal::factory( array( 'class' => 'NullFileJournal' ), $this->name );
 110                 $this->readOnly = isset( $config['readOnly'] )
 111                         ? (string)$config['readOnly']
 112                         : '';
 113                 $this->parallelize = isset( $config['parallelize'] )
 114                         ? (string)$config['parallelize']
 115                         : 'off';
 116                 $this->concurrency = isset( $config['concurrency'] )
 117                         ? (int)$config['concurrency']
 118                         : 50;
 119         }
 120
 121         /**
 122          * Get the unique backend name.
 123          * We may have multiple different backends of the same type.
 124          * For example, we can have two Swift backends using different proxies.
 125          *
 126          * @return string
 127          */
 128         final public function getName() {
 129                 return $this->name;
 130         }
 131
 132         /**
 133          * Check if this backend is read-only
 134          *
 135          * @return bool
 136          */
 137         final public function isReadOnly() {
 138                 return ( $this->readOnly != '' );
 139         }
 140
 141         /**
 142          * Get an explanatory message if this backend is read-only
 143          *
 144          * @return string|bool Returns false if the backend is not read-only
 145          */
 146         final public function getReadOnlyReason() {
 147                 return ( $this->readOnly != '' ) ? $this->readOnly : false;
 148         }
 149
 150         /**
 151          * This is the main entry point into the backend for write operations.
 152          * Callers supply an ordered list of operations to perform as a transaction.
 153          * Files will be locked, the stat cache cleared, and then the operations attempted.
 154          * If any serious errors occur, all attempted operations will be rolled back.
 155          *
 156          * $ops is an array of arrays. The outer array holds a list of operations.
 157          * Each inner array is a set of key value pairs that specify an operation.
 158          *
 159          * Supported operations and their parameters. The supported actions are:
 160          *  - create
 161          *  - store
 162          *  - copy
 163          *  - move
 164          *  - delete
 165          *  - null
 166          *
 167          * a) Create a new file in storage with the contents of a string
 168          * @code
 169          *     array(
 170          *         'op'                  => 'create',
 171          *         'dst'                 => <storage path>,
 172          *         'content'             => <string of new file contents>,
 173          *         'overwrite'           => <boolean>,
 174          *         'overwriteSame'       => <boolean>
 175          *     );
 176          * @endcode
 177          *
 178          * b) Copy a file system file into storage
 179          * @code
 180          *     array(
 181          *         'op'                  => 'store',
 182          *         'src'                 => <file system path>,
 183          *         'dst'                 => <storage path>,
 184          *         'overwrite'           => <boolean>,
 185          *         'overwriteSame'       => <boolean>
 186          *     )
 187          * @endcode
 188          *
 189          * c) Copy a file within storage
 190          * @code
 191          *     array(
 192          *         'op'                  => 'copy',
 193          *         'src'                 => <storage path>,
 194          *         'dst'                 => <storage path>,
 195          *         'overwrite'           => <boolean>,
 196          *         'overwriteSame'       => <boolean>
 197          *     )
 198          * @endcode
 199          *
 200          * d) Move a file within storage
 201          * @code
 202          *     array(
 203          *         'op'                  => 'move',
 204          *         'src'                 => <storage path>,
 205          *         'dst'                 => <storage path>,
 206          *         'overwrite'           => <boolean>,
 207          *         'overwriteSame'       => <boolean>
 208          *     )
 209          * @endcode
 210          *
 211          * e) Delete a file within storage
 212          * @code
 213          *     array(
 214          *         'op'                  => 'delete',
 215          *         'src'                 => <storage path>,
 216          *         'ignoreMissingSource' => <boolean>
 217          *     )
 218          * @endcode
 219          *
 220          * f) Do nothing (no-op)
 221          * @code
 222          *     array(
 223          *         'op'                  => 'null',
 224          *     )
 225          * @endcode
 226          *
 227          * Boolean flags for operations (operation-specific):
 228          *   - ignoreMissingSource : The operation will simply succeed and do
 229          *                           nothing if the source file does not exist.
 230          *   - overwrite           : Any destination file will be overwritten.
 231          *   - overwriteSame       : An error will not be given if a file already
 232          *                           exists at the destination that has the same
 233          *                           contents as the new contents to be written there.
 234          *
 235          * $opts is an associative of boolean flags, including:
 236          *   - force               : Operation precondition errors no longer trigger an abort.
 237          *                           Any remaining operations are still attempted. Unexpected
 238          *                           failures may still cause remaning operations to be aborted.
 239          *   - nonLocking          : No locks are acquired for the operations.
 240          *                           This can increase performance for non-critical writes.
 241          *                           This has no effect unless the 'force' flag is set.
 242          *   - allowStale          : Don't require the latest available data.
 243          *                           This can increase performance for non-critical writes.
 244          *                           This has no effect unless the 'force' flag is set.
 245          *   - nonJournaled        : Don't log this operation batch in the file journal.
 246          *                           This limits the ability of recovery scripts.
 247          *   - parallelize         : Try to do operations in parallel when possible.
 248          *   - bypassReadOnly      : Allow writes in read-only mode (since 1.20).
 249          *
 250          * @remarks Remarks on locking:
 251          * File system paths given to operations should refer to files that are
 252          * already locked or otherwise safe from modification from other processes.
 253          * Normally these files will be new temp files, which should be adequate.
 254          *
 255          * @par Return value:
 256          *
 257          * This returns a Status, which contains all warnings and fatals that occured
 258          * during the operation. The 'failCount', 'successCount', and 'success' members
 259          * will reflect each operation attempted.
 260          *
 261          * The status will be "OK" unless:
 262          *   - a) unexpected operation errors occurred (network partitions, disk full...)
 263          *   - b) significant operation errors occured and 'force' was not set
 264          *
 265          * @param $ops Array List of operations to execute in order
 266          * @param $opts Array Batch operation options
 267          * @return Status
 268          */
 269         final public function doOperations( array $ops, array $opts = array() ) {
 270                 if ( empty( $opts['bypassReadOnly'] ) && $this->isReadOnly() ) {
 271                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 272                 }
 273                 if ( empty( $opts['force'] ) ) { // sanity
 274                         unset( $opts['nonLocking'] );
 275                         unset( $opts['allowStale'] );
 276                 }
 277                 $opts['concurrency'] = 1; // off
 278                 if ( $this->parallelize === 'implicit' ) {
 279                         if ( !isset( $opts['parallelize'] ) || $opts['parallelize'] ) {
 280                                 $opts['concurrency'] = $this->concurrency;
 281                         }
 282                 } elseif ( $this->parallelize === 'explicit' ) {
 283                         if ( !empty( $opts['parallelize'] ) ) {
 284                                 $opts['concurrency'] = $this->concurrency;
 285                         }
 286                 }
 287                 return $this->doOperationsInternal( $ops, $opts );
 288         }
 289
 290         /**
 291          * @see FileBackend::doOperations()
 292          */
 293         abstract protected function doOperationsInternal( array $ops, array $opts );
 294
 295         /**
 296          * Same as doOperations() except it takes a single operation.
 297          * If you are doing a batch of operations that should either
 298          * all succeed or all fail, then use that function instead.
 299          *
 300          * @see FileBackend::doOperations()
 301          *
 302          * @param $op Array Operation
 303          * @param $opts Array Operation options
 304          * @return Status
 305          */
 306         final public function doOperation( array $op, array $opts = array() ) {
 307                 return $this->doOperations( array( $op ), $opts );
 308         }
 309
 310         /**
 311          * Performs a single create operation.
 312          * This sets $params['op'] to 'create' and passes it to doOperation().
 313          *
 314          * @see FileBackend::doOperation()
 315          *
 316          * @param $params Array Operation parameters
 317          * @param $opts Array Operation options
 318          * @return Status
 319          */
 320         final public function create( array $params, array $opts = array() ) {
 321                 return $this->doOperation( array( 'op' => 'create' ) + $params, $opts );
 322         }
 323
 324         /**
 325          * Performs a single store operation.
 326          * This sets $params['op'] to 'store' and passes it to doOperation().
 327          *
 328          * @see FileBackend::doOperation()
 329          *
 330          * @param $params Array Operation parameters
 331          * @param $opts Array Operation options
 332          * @return Status
 333          */
 334         final public function store( array $params, array $opts = array() ) {
 335                 return $this->doOperation( array( 'op' => 'store' ) + $params, $opts );
 336         }
 337
 338         /**
 339          * Performs a single copy operation.
 340          * This sets $params['op'] to 'copy' and passes it to doOperation().
 341          *
 342          * @see FileBackend::doOperation()
 343          *
 344          * @param $params Array Operation parameters
 345          * @param $opts Array Operation options
 346          * @return Status
 347          */
 348         final public function copy( array $params, array $opts = array() ) {
 349                 return $this->doOperation( array( 'op' => 'copy' ) + $params, $opts );
 350         }
 351
 352         /**
 353          * Performs a single move operation.
 354          * This sets $params['op'] to 'move' and passes it to doOperation().
 355          *
 356          * @see FileBackend::doOperation()
 357          *
 358          * @param $params Array Operation parameters
 359          * @param $opts Array Operation options
 360          * @return Status
 361          */
 362         final public function move( array $params, array $opts = array() ) {
 363                 return $this->doOperation( array( 'op' => 'move' ) + $params, $opts );
 364         }
 365
 366         /**
 367          * Performs a single delete operation.
 368          * This sets $params['op'] to 'delete' and passes it to doOperation().
 369          *
 370          * @see FileBackend::doOperation()
 371          *
 372          * @param $params Array Operation parameters
 373          * @param $opts Array Operation options
 374          * @return Status
 375          */
 376         final public function delete( array $params, array $opts = array() ) {
 377                 return $this->doOperation( array( 'op' => 'delete' ) + $params, $opts );
 378         }
 379
 380         /**
 381          * Perform a set of independent file operations on some files.
 382          *
 383          * This does no locking, nor journaling, and possibly no stat calls.
 384          * Any destination files that already exist will be overwritten.
 385          * This should *only* be used on non-original files, like cache files.
 386          *
 387          * Supported operations and their parameters:
 388          *  - create
 389          *  - store
 390          *  - copy
 391          *  - move
 392          *  - delete
 393          *  - null
 394          *
 395          * a) Create a new file in storage with the contents of a string
 396          * @code
 397          *     array(
 398          *         'op'                  => 'create',
 399          *         'dst'                 => <storage path>,
 400          *         'content'             => <string of new file contents>
 401          *     )
 402          * @endcode
 403          * b) Copy a file system file into storage
 404          * @code
 405          *     array(
 406          *         'op'                  => 'store',
 407          *         'src'                 => <file system path>,
 408          *         'dst'                 => <storage path>
 409          *     )
 410          * @endcode
 411          * c) Copy a file within storage
 412          * @code
 413          *     array(
 414          *         'op'                  => 'copy',
 415          *         'src'                 => <storage path>,
 416          *         'dst'                 => <storage path>
 417          *     )
 418          * @endcode
 419          * d) Move a file within storage
 420          * @code
 421          *     array(
 422          *         'op'                  => 'move',
 423          *         'src'                 => <storage path>,
 424          *         'dst'                 => <storage path>
 425          *     )
 426          * @endcode
 427          * e) Delete a file within storage
 428          * @code
 429          *     array(
 430          *         'op'                  => 'delete',
 431          *         'src'                 => <storage path>,
 432          *         'ignoreMissingSource' => <boolean>
 433          *     )
 434          * @endcode
 435          * f) Do nothing (no-op)
 436          * @code
 437          *     array(
 438          *         'op'                  => 'null',
 439          *     )
 440          * @endcode
 441          *
 442          * @par Boolean flags for operations (operation-specific):
 443          *   - ignoreMissingSource : The operation will simply succeed and do
 444          *                           nothing if the source file does not exist.
 445          *
 446          * $opts is an associative of boolean flags, including:
 447          *   - bypassReadOnly      : Allow writes in read-only mode (since 1.20)
 448          *
 449          * @par Return value:
 450          * This returns a Status, which contains all warnings and fatals that occured
 451          * during the operation. The 'failCount', 'successCount', and 'success' members
 452          * will reflect each operation attempted for the given files. The status will be
 453          * considered "OK" as long as no fatal errors occured.
 454          *
 455          * @param $ops Array Set of operations to execute
 456          * @param $opts Array Batch operation options
 457          * @return Status
 458          * @since 1.20
 459          */
 460         final public function doQuickOperations( array $ops, array $opts = array() ) {
 461                 if ( empty( $opts['bypassReadOnly'] ) && $this->isReadOnly() ) {
 462                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 463                 }
 464                 foreach ( $ops as &$op ) {
 465                         $op['overwrite'] = true; // avoids RTTs in key/value stores
 466                 }
 467                 return $this->doQuickOperationsInternal( $ops );
 468         }
 469
 470         /**
 471          * @see FileBackend::doQuickOperations()
 472          * @since 1.20
 473          */
 474         abstract protected function doQuickOperationsInternal( array $ops );
 475
 476         /**
 477          * Same as doQuickOperations() except it takes a single operation.
 478          * If you are doing a batch of operations, then use that function instead.
 479          *
 480          * @see FileBackend::doQuickOperations()
 481          *
 482          * @param $op Array Operation
 483          * @return Status
 484          * @since 1.20
 485          */
 486         final public function doQuickOperation( array $op ) {
 487                 return $this->doQuickOperations( array( $op ) );
 488         }
 489
 490         /**
 491          * Performs a single quick create operation.
 492          * This sets $params['op'] to 'create' and passes it to doQuickOperation().
 493          *
 494          * @see FileBackend::doQuickOperation()
 495          *
 496          * @param $params Array Operation parameters
 497          * @return Status
 498          * @since 1.20
 499          */
 500         final public function quickCreate( array $params ) {
 501                 return $this->doQuickOperation( array( 'op' => 'create' ) + $params );
 502         }
 503
 504         /**
 505          * Performs a single quick store operation.
 506          * This sets $params['op'] to 'store' and passes it to doQuickOperation().
 507          *
 508          * @see FileBackend::doQuickOperation()
 509          *
 510          * @param $params Array Operation parameters
 511          * @return Status
 512          * @since 1.20
 513          */
 514         final public function quickStore( array $params ) {
 515                 return $this->doQuickOperation( array( 'op' => 'store' ) + $params );
 516         }
 517
 518         /**
 519          * Performs a single quick copy operation.
 520          * This sets $params['op'] to 'copy' and passes it to doQuickOperation().
 521          *
 522          * @see FileBackend::doQuickOperation()
 523          *
 524          * @param $params Array Operation parameters
 525          * @return Status
 526          * @since 1.20
 527          */
 528         final public function quickCopy( array $params ) {
 529                 return $this->doQuickOperation( array( 'op' => 'copy' ) + $params );
 530         }
 531
 532         /**
 533          * Performs a single quick move operation.
 534          * This sets $params['op'] to 'move' and passes it to doQuickOperation().
 535          *
 536          * @see FileBackend::doQuickOperation()
 537          *
 538          * @param $params Array Operation parameters
 539          * @return Status
 540          * @since 1.20
 541          */
 542         final public function quickMove( array $params ) {
 543                 return $this->doQuickOperation( array( 'op' => 'move' ) + $params );
 544         }
 545
 546         /**
 547          * Performs a single quick delete operation.
 548          * This sets $params['op'] to 'delete' and passes it to doQuickOperation().
 549          *
 550          * @see FileBackend::doQuickOperation()
 551          *
 552          * @param $params Array Operation parameters
 553          * @return Status
 554          * @since 1.20
 555          */
 556         final public function quickDelete( array $params ) {
 557                 return $this->doQuickOperation( array( 'op' => 'delete' ) + $params );
 558         }
 559
 560         /**
 561          * Concatenate a list of storage files into a single file system file.
 562          * The target path should refer to a file that is already locked or
 563          * otherwise safe from modification from other processes. Normally,
 564          * the file will be a new temp file, which should be adequate.
 565          *
 566          * @param $params Array Operation parameters
 567          * $params include:
 568          *   - srcs : ordered source storage paths (e.g. chunk1, chunk2, ...)
 569          *   - dst  : file system path to 0-byte temp file
 570          * @return Status
 571          */
 572         abstract public function concatenate( array $params );
 573
 574         /**
 575          * Prepare a storage directory for usage.
 576          * This will create any required containers and parent directories.
 577          * Backends using key/value stores only need to create the container.
 578          *
 579          * The 'noAccess' and 'noListing' parameters works the same as in secure(),
 580          * except they are only applied *if* the directory/container had to be created.
 581          * These flags should always be set for directories that have private files.
 582          *
 583          * @param $params Array
 584          * $params include:
 585          *   - dir            : storage directory
 586          *   - noAccess       : try to deny file access (since 1.20)
 587          *   - noListing      : try to deny file listing (since 1.20)
 588          *   - bypassReadOnly : allow writes in read-only mode (since 1.20)
 589          * @return Status
 590          */
 591         final public function prepare( array $params ) {
 592                 if ( empty( $params['bypassReadOnly'] ) && $this->isReadOnly() ) {
 593                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 594                 }
 595                 return $this->doPrepare( $params );
 596         }
 597
 598         /**
 599          * @see FileBackend::prepare()
 600          */
 601         abstract protected function doPrepare( array $params );
 602
 603         /**
 604          * Take measures to block web access to a storage directory and
 605          * the container it belongs to. FS backends might add .htaccess
 606          * files whereas key/value store backends might revoke container
 607          * access to the storage user representing end-users in web requests.
 608          * This is not guaranteed to actually do anything.
 609          *
 610          * @param $params Array
 611          * $params include:
 612          *   - dir            : storage directory
 613          *   - noAccess       : try to deny file access
 614          *   - noListing      : try to deny file listing
 615          *   - bypassReadOnly : allow writes in read-only mode (since 1.20)
 616          * @return Status
 617          */
 618         final public function secure( array $params ) {
 619                 if ( empty( $params['bypassReadOnly'] ) && $this->isReadOnly() ) {
 620                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 621                 }
 622                 return $this->doSecure( $params );
 623         }
 624
 625         /**
 626          * @see FileBackend::secure()
 627          */
 628         abstract protected function doSecure( array $params );
 629
 630         /**
 631          * Remove measures to block web access to a storage directory and
 632          * the container it belongs to. FS backends might remove .htaccess
 633          * files whereas key/value store backends might grant container
 634          * access to the storage user representing end-users in web requests.
 635          * This essentially can undo the result of secure() calls.
 636          *
 637          * @param $params Array
 638          * $params include:
 639          *   - dir            : storage directory
 640          *   - access         : try to allow file access
 641          *   - listing        : try to allow file listing
 642          *   - bypassReadOnly : allow writes in read-only mode (since 1.20)
 643          * @return Status
 644          * @since 1.20
 645          */
 646         final public function publish( array $params ) {
 647                 if ( empty( $params['bypassReadOnly'] ) && $this->isReadOnly() ) {
 648                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 649                 }
 650                 return $this->doPublish( $params );
 651         }
 652
 653         /**
 654          * @see FileBackend::publish()
 655          */
 656         abstract protected function doPublish( array $params );
 657
 658         /**
 659          * Delete a storage directory if it is empty.
 660          * Backends using key/value stores may do nothing unless the directory
 661          * is that of an empty container, in which case it should be deleted.
 662          *
 663          * @param $params Array
 664          * $params include:
 665          *   - dir            : storage directory
 666          *   - recursive      : recursively delete empty subdirectories first (since 1.20)
 667          *   - bypassReadOnly : allow writes in read-only mode (since 1.20)
 668          * @return Status
 669          */
 670         final public function clean( array $params ) {
 671                 if ( empty( $params['bypassReadOnly'] ) && $this->isReadOnly() ) {
 672                         return Status::newFatal( 'backend-fail-readonly', $this->name, $this->readOnly );
 673                 }
 674                 return $this->doClean( $params );
 675         }
 676
 677         /**
 678          * @see FileBackend::clean()
 679          */
 680         abstract protected function doClean( array $params );
 681
 682         /**
 683          * Check if a file exists at a storage path in the backend.
 684          * This returns false if only a directory exists at the path.
 685          *
 686          * @param $params Array
 687          * $params include:
 688          *   - src    : source storage path
 689          *   - latest : use the latest available data
 690          * @return bool|null Returns null on failure
 691          */
 692         abstract public function fileExists( array $params );
 693
 694         /**
 695          * Get the last-modified timestamp of the file at a storage path.
 696          *
 697          * @param $params Array
 698          * $params include:
 699          *   - src    : source storage path
 700          *   - latest : use the latest available data
 701          * @return string|bool TS_MW timestamp or false on failure
 702          */
 703         abstract public function getFileTimestamp( array $params );
 704
 705         /**
 706          * Get the contents of a file at a storage path in the backend.
 707          * This should be avoided for potentially large files.
 708          *
 709          * @param $params Array
 710          * $params include:
 711          *   - src    : source storage path
 712          *   - latest : use the latest available data
 713          * @return string|bool Returns false on failure
 714          */
 715         abstract public function getFileContents( array $params );
 716
 717         /**
 718          * Get the size (bytes) of a file at a storage path in the backend.
 719          *
 720          * @param $params Array
 721          * $params include:
 722          *   - src    : source storage path
 723          *   - latest : use the latest available data
 724          * @return integer|bool Returns false on failure
 725          */
 726         abstract public function getFileSize( array $params );
 727
 728         /**
 729          * Get quick information about a file at a storage path in the backend.
 730          * If the file does not exist, then this returns false.
 731          * Otherwise, the result is an associative array that includes:
 732          *   - mtime  : the last-modified timestamp (TS_MW)
 733          *   - size   : the file size (bytes)
 734          * Additional values may be included for internal use only.
 735          *
 736          * @param $params Array
 737          * $params include:
 738          *   - src    : source storage path
 739          *   - latest : use the latest available data
 740          * @return Array|bool|null Returns null on failure
 741          */
 742         abstract public function getFileStat( array $params );
 743
 744         /**
 745          * Get a SHA-1 hash of the file at a storage path in the backend.
 746          *
 747          * @param $params Array
 748          * $params include:
 749          *   - src    : source storage path
 750          *   - latest : use the latest available data
 751          * @return string|bool Hash string or false on failure
 752          */
 753         abstract public function getFileSha1Base36( array $params );
 754
 755         /**
 756          * Get the properties of the file at a storage path in the backend.
 757          * Returns FSFile::placeholderProps() on failure.
 758          *
 759          * @param $params Array
 760          * $params include:
 761          *   - src    : source storage path
 762          *   - latest : use the latest available data
 763          * @return Array
 764          */
 765         abstract public function getFileProps( array $params );
 766
 767         /**
 768          * Stream the file at a storage path in the backend.
 769          * If the file does not exists, a 404 error will be given.
 770          * Appropriate HTTP headers (Status, Content-Type, Content-Length)
 771          * must be sent if streaming began, while none should be sent otherwise.
 772          * Implementations should flush the output buffer before sending data.
 773          *
 774          * @param $params Array
 775          * $params include:
 776          *   - src     : source storage path
 777          *   - headers : additional HTTP headers to send on success
 778          *   - latest  : use the latest available data
 779          * @return Status
 780          */
 781         abstract public function streamFile( array $params );
 782
 783         /**
 784          * Returns a file system file, identical to the file at a storage path.
 785          * The file returned is either:
 786          *   - a) A local copy of the file at a storage path in the backend.
 787          *        The temporary copy will have the same extension as the source.
 788          *   - b) An original of the file at a storage path in the backend.
 789          * Temporary files may be purged when the file object falls out of scope.
 790          *
 791          * Write operations should *never* be done on this file as some backends
 792          * may do internal tracking or may be instances of FileBackendMultiWrite.
 793          * In that later case, there are copies of the file that must stay in sync.
 794          * Additionally, further calls to this function may return the same file.
 795          *
 796          * @param $params Array
 797          * $params include:
 798          *   - src    : source storage path
 799          *   - latest : use the latest available data
 800          * @return FSFile|null Returns null on failure
 801          */
 802         abstract public function getLocalReference( array $params );
 803
 804         /**
 805          * Get a local copy on disk of the file at a storage path in the backend.
 806          * The temporary copy will have the same file extension as the source.
 807          * Temporary files may be purged when the file object falls out of scope.
 808          *
 809          * @param $params Array
 810          * $params include:
 811          *   - src    : source storage path
 812          *   - latest : use the latest available data
 813          * @return TempFSFile|null Returns null on failure
 814          */
 815         abstract public function getLocalCopy( array $params );
 816
 817         /**
 818          * Check if a directory exists at a given storage path.
 819          * Backends using key/value stores will check if the path is a
 820          * virtual directory, meaning there are files under the given directory.
 821          *
 822          * Storage backends with eventual consistency might return stale data.
 823          *
 824          * @param $params array
 825          * $params include:
 826          *   - dir : storage directory
 827          * @return bool|null Returns null on failure
 828          * @since 1.20
 829          */
 830         abstract public function directoryExists( array $params );
 831
 832         /**
 833          * Get an iterator to list *all* directories under a storage directory.
 834          * If the directory is of the form "mwstore://backend/container",
 835          * then all directories in the container should be listed.
 836          * If the directory is of form "mwstore://backend/container/dir",
 837          * then all directories directly under that directory should be listed.
 838          * Results should be storage directories relative to the given directory.
 839          *
 840          * Storage backends with eventual consistency might return stale data.
 841          *
 842          * @param $params array
 843          * $params include:
 844          *   - dir     : storage directory
 845          *   - topOnly : only return direct child dirs of the directory
 846          * @return Traversable|Array|null Returns null on failure
 847          * @since 1.20
 848          */
 849         abstract public function getDirectoryList( array $params );
 850
 851         /**
 852          * Same as FileBackend::getDirectoryList() except only lists
 853          * directories that are immediately under the given directory.
 854          *
 855          * Storage backends with eventual consistency might return stale data.
 856          *
 857          * @param $params array
 858          * $params include:
 859          *   - dir : storage directory
 860          * @return Traversable|Array|null Returns null on failure
 861          * @since 1.20
 862          */
 863         final public function getTopDirectoryList( array $params ) {
 864                 return $this->getDirectoryList( array( 'topOnly' => true ) + $params );
 865         }
 866
 867         /**
 868          * Get an iterator to list *all* stored files under a storage directory.
 869          * If the directory is of the form "mwstore://backend/container",
 870          * then all files in the container should be listed.
 871          * If the directory is of form "mwstore://backend/container/dir",
 872          * then all files under that directory should be listed.
 873          * Results should be storage paths relative to the given directory.
 874          *
 875          * Storage backends with eventual consistency might return stale data.
 876          *
 877          * @param $params array
 878          * $params include:
 879          *   - dir     : storage directory
 880          *   - topOnly : only return direct child files of the directory (since 1.20)
 881          * @return Traversable|Array|null Returns null on failure
 882          */
 883         abstract public function getFileList( array $params );
 884
 885         /**
 886          * Same as FileBackend::getFileList() except only lists
 887          * files that are immediately under the given directory.
 888          *
 889          * Storage backends with eventual consistency might return stale data.
 890          *
 891          * @param $params array
 892          * $params include:
 893          *   - dir : storage directory
 894          * @return Traversable|Array|null Returns null on failure
 895          * @since 1.20
 896          */
 897         final public function getTopFileList( array $params ) {
 898                 return $this->getFileList( array( 'topOnly' => true ) + $params );
 899         }
 900
 901         /**
 902          * Invalidate any in-process file existence and property cache.
 903          * If $paths is given, then only the cache for those files will be cleared.
 904          *
 905          * @param $paths Array Storage paths (optional)
 906          * @return void
 907          */
 908         public function clearCache( array $paths = null ) {}
 909
 910         /**
 911          * Lock the files at the given storage paths in the backend.
 912          * This will either lock all the files or none (on failure).
 913          *
 914          * Callers should consider using getScopedFileLocks() instead.
 915          *
 916          * @param $paths Array Storage paths
 917          * @param $type integer LockManager::LOCK_* constant
 918          * @return Status
 919          */
 920         final public function lockFiles( array $paths, $type ) {
 921                 return $this->lockManager->lock( $paths, $type );
 922         }
 923
 924         /**
 925          * Unlock the files at the given storage paths in the backend.
 926          *
 927          * @param $paths Array Storage paths
 928          * @param $type integer LockManager::LOCK_* constant
 929          * @return Status
 930          */
 931         final public function unlockFiles( array $paths, $type ) {
 932                 return $this->lockManager->unlock( $paths, $type );
 933         }
 934
 935         /**
 936          * Lock the files at the given storage paths in the backend.
 937          * This will either lock all the files or none (on failure).
 938          * On failure, the status object will be updated with errors.
 939          *
 940          * Once the return value goes out scope, the locks will be released and
 941          * the status updated. Unlock fatals will not change the status "OK" value.
 942          *
 943          * @param $paths Array Storage paths
 944          * @param $type integer LockManager::LOCK_* constant
 945          * @param $status Status Status to update on lock/unlock
 946          * @return ScopedLock|null Returns null on failure
 947          */
 948         final public function getScopedFileLocks( array $paths, $type, Status $status ) {
 949                 return ScopedLock::factory( $this->lockManager, $paths, $type, $status );
 950         }
 951
 952         /**
 953          * Get an array of scoped locks needed for a batch of file operations.
 954          *
 955          * Normally, FileBackend::doOperations() handles locking, unless
 956          * the 'nonLocking' param is passed in. This function is useful if you
 957          * want the files to be locked for a broader scope than just when the
 958          * files are changing. For example, if you need to update DB metadata,
 959          * you may want to keep the files locked until finished.
 960          *
 961          * @see FileBackend::doOperations()
 962          *
 963          * @param $ops Array List of file operations to FileBackend::doOperations()
 964          * @param $status Status Status to update on lock/unlock
 965          * @return Array List of ScopedFileLocks or null values
 966          * @since 1.20
 967          */
 968         abstract public function getScopedLocksForOps( array $ops, Status $status );
 969
 970         /**
 971          * Get the root storage path of this backend.
 972          * All container paths are "subdirectories" of this path.
 973          *
 974          * @return string Storage path
 975          * @since 1.20
 976          */
 977         final public function getRootStoragePath() {
 978                 return "mwstore://{$this->name}";
 979         }
 980
 981         /**
 982          * Get the file journal object for this backend
 983          *
 984          * @return FileJournal
 985          */
 986         final public function getJournal() {
 987                 return $this->fileJournal;
 988         }
 989
 990         /**
 991          * Check if a given path is a "mwstore://" path.
 992          * This does not do any further validation or any existence checks.
 993          *
 994          * @param $path string
 995          * @return bool
 996          */
 997         final public static function isStoragePath( $path ) {
 998                 return ( strpos( $path, 'mwstore://' ) === 0 );
 999         }
1000
1001         /**
1002          * Split a storage path into a backend name, a container name,
1003          * and a relative file path. The relative path may be the empty string.
1004          * This does not do any path normalization or traversal checks.
1005          *
1006          * @param $storagePath string
1007          * @return Array (backend, container, rel object) or (null, null, null)
1008          */
1009         final public static function splitStoragePath( $storagePath ) {
1010                 if ( self::isStoragePath( $storagePath ) ) {
1011                         // Remove the "mwstore://" prefix and split the path
1012                         $parts = explode( '/', substr( $storagePath, 10 ), 3 );
1013                         if ( count( $parts ) >= 2 && $parts[0] != '' && $parts[1] != '' ) {
1014                                 if ( count( $parts ) == 3 ) {
1015                                         return $parts; // e.g. "backend/container/path"
1016                                 } else {
1017                                         return array( $parts[0], $parts[1], '' ); // e.g. "backend/container"
1018                                 }
1019                         }
1020                 }
1021                 return array( null, null, null );
1022         }
1023
1024         /**
1025          * Normalize a storage path by cleaning up directory separators.
1026          * Returns null if the path is not of the format of a valid storage path.
1027          *
1028          * @param $storagePath string
1029          * @return string|null
1030          */
1031         final public static function normalizeStoragePath( $storagePath ) {
1032                 list( $backend, $container, $relPath ) = self::splitStoragePath( $storagePath );
1033                 if ( $relPath !== null ) { // must be for this backend
1034                         $relPath = self::normalizeContainerPath( $relPath );
1035                         if ( $relPath !== null ) {
1036                                 return ( $relPath != '' )
1037                                         ? "mwstore://{$backend}/{$container}/{$relPath}"
1038                                         : "mwstore://{$backend}/{$container}";
1039                         }
1040                 }
1041                 return null;
1042         }
1043
1044         /**
1045          * Get the parent storage directory of a storage path.
1046          * This returns a path like "mwstore://backend/container",
1047          * "mwstore://backend/container/...", or null if there is no parent.
1048          *
1049          * @param $storagePath string
1050          * @return string|null
1051          */
1052         final public static function parentStoragePath( $storagePath ) {
1053                 $storagePath = dirname( $storagePath );
1054                 list( $b, $cont, $rel ) = self::splitStoragePath( $storagePath );
1055                 return ( $rel === null ) ? null : $storagePath;
1056         }
1057
1058         /**
1059          * Get the final extension from a storage or FS path
1060          *
1061          * @param $path string
1062          * @return string
1063          */
1064         final public static function extensionFromPath( $path ) {
1065                 $i = strrpos( $path, '.' );
1066                 return strtolower( $i ? substr( $path, $i + 1 ) : '' );
1067         }
1068
1069         /**
1070          * Check if a relative path has no directory traversals
1071          *
1072          * @param $path string
1073          * @return bool
1074          * @since 1.20
1075          */
1076         final public static function isPathTraversalFree( $path ) {
1077                 return ( self::normalizeContainerPath( $path ) !== null );
1078         }
1079
1080         /**
1081          * Validate and normalize a relative storage path.
1082          * Null is returned if the path involves directory traversal.
1083          * Traversal is insecure for FS backends and broken for others.
1084          *
1085          * This uses the same traversal protection as Title::secureAndSplit().
1086          *
1087          * @param $path string Storage path relative to a container
1088          * @return string|null
1089          */
1090         final protected static function normalizeContainerPath( $path ) {
1091                 // Normalize directory separators
1092                 $path = strtr( $path, '\\', '/' );
1093                 // Collapse any consecutive directory separators
1094                 $path = preg_replace( '![/]{2,}!', '/', $path );
1095                 // Remove any leading directory separator
1096                 $path = ltrim( $path, '/' );
1097                 // Use the same traversal protection as Title::secureAndSplit()
1098                 if ( strpos( $path, '.' ) !== false ) {
1099                         if (
1100                                 $path === '.' ||
1101                                 $path === '..' ||
1102                                 strpos( $path, './' ) === 0 ||
1103                                 strpos( $path, '../' ) === 0 ||
1104                                 strpos( $path, '/./' ) !== false ||
1105                                 strpos( $path, '/../' ) !== false
1106                         ) {
1107                                 return null;
1108                         }
1109                 }
1110                 return $path;
1111         }
1112 }