sdk/python/arvados/arvfile.py

   1 import functools
   2 import os
   3 import zlib
   4 import bz2
   5 import config
   6 import hashlib
   7 import threading
   8 import Queue
   9 import copy
  10 import errno
  11 import re
  12 import logging
  13 import collections
  14 import uuid
  15
  16 from .errors import KeepWriteError, AssertionError, ArgumentError
  17 from .keep import KeepLocator
  18 from ._normalize_stream import normalize_stream
  19 from ._ranges import locators_and_ranges, replace_range, Range
  20 from .retry import retry_method
  21
  22 MOD = "mod"
  23 WRITE = "write"
  24
  25 _logger = logging.getLogger('arvados.arvfile')
  26
  27 def split(path):
  28     """split(path) -> streamname, filename
  29
  30     Separate the stream name and file name in a /-separated stream path and
  31     return a tuple (stream_name, file_name).  If no stream name is available,
  32     assume '.'.
  33
  34     """
  35     try:
  36         stream_name, file_name = path.rsplit('/', 1)
  37     except ValueError:  # No / in string
  38         stream_name, file_name = '.', path
  39     return stream_name, file_name
  40
  41 class _FileLikeObjectBase(object):
  42     def __init__(self, name, mode):
  43         self.name = name
  44         self.mode = mode
  45         self.closed = False
  46
  47     @staticmethod
  48     def _before_close(orig_func):
  49         @functools.wraps(orig_func)
  50         def before_close_wrapper(self, *args, **kwargs):
  51             if self.closed:
  52                 raise ValueError("I/O operation on closed stream file")
  53             return orig_func(self, *args, **kwargs)
  54         return before_close_wrapper
  55
  56     def __enter__(self):
  57         return self
  58
  59     def __exit__(self, exc_type, exc_value, traceback):
  60         try:
  61             self.close()
  62         except Exception:
  63             if exc_type is None:
  64                 raise
  65
  66     def close(self):
  67         self.closed = True
  68
  69
  70 class ArvadosFileReaderBase(_FileLikeObjectBase):
  71     def __init__(self, name, mode, num_retries=None):
  72         super(ArvadosFileReaderBase, self).__init__(name, mode)
  73         self._filepos = 0L
  74         self.num_retries = num_retries
  75         self._readline_cache = (None, None)
  76
  77     def __iter__(self):
  78         while True:
  79             data = self.readline()
  80             if not data:
  81                 break
  82             yield data
  83
  84     def decompressed_name(self):
  85         return re.sub('\.(bz2|gz)$', '', self.name)
  86
  87     @_FileLikeObjectBase._before_close
  88     def seek(self, pos, whence=os.SEEK_SET):
  89         if whence == os.SEEK_CUR:
  90             pos += self._filepos
  91         elif whence == os.SEEK_END:
  92             pos += self.size()
  93         self._filepos = min(max(pos, 0L), self.size())
  94
  95     def tell(self):
  96         return self._filepos
  97
  98     @_FileLikeObjectBase._before_close
  99     @retry_method
 100     def readall(self, size=2**20, num_retries=None):
 101         while True:
 102             data = self.read(size, num_retries=num_retries)
 103             if data == '':
 104                 break
 105             yield data
 106
 107     @_FileLikeObjectBase._before_close
 108     @retry_method
 109     def readline(self, size=float('inf'), num_retries=None):
 110         cache_pos, cache_data = self._readline_cache
 111         if self.tell() == cache_pos:
 112             data = [cache_data]
 113             self._filepos += len(cache_data)
 114         else:
 115             data = ['']
 116         data_size = len(data[-1])
 117         while (data_size < size) and ('\n' not in data[-1]):
 118             next_read = self.read(2 ** 20, num_retries=num_retries)
 119             if not next_read:
 120                 break
 121             data.append(next_read)
 122             data_size += len(next_read)
 123         data = ''.join(data)
 124         try:
 125             nextline_index = data.index('\n') + 1
 126         except ValueError:
 127             nextline_index = len(data)
 128         nextline_index = min(nextline_index, size)
 129         self._filepos -= len(data) - nextline_index
 130         self._readline_cache = (self.tell(), data[nextline_index:])
 131         return data[:nextline_index]
 132
 133     @_FileLikeObjectBase._before_close
 134     @retry_method
 135     def decompress(self, decompress, size, num_retries=None):
 136         for segment in self.readall(size, num_retries=num_retries):
 137             data = decompress(segment)
 138             if data:
 139                 yield data
 140
 141     @_FileLikeObjectBase._before_close
 142     @retry_method
 143     def readall_decompressed(self, size=2**20, num_retries=None):
 144         self.seek(0)
 145         if self.name.endswith('.bz2'):
 146             dc = bz2.BZ2Decompressor()
 147             return self.decompress(dc.decompress, size,
 148                                    num_retries=num_retries)
 149         elif self.name.endswith('.gz'):
 150             dc = zlib.decompressobj(16+zlib.MAX_WBITS)
 151             return self.decompress(lambda segment: dc.decompress(dc.unconsumed_tail + segment),
 152                                    size, num_retries=num_retries)
 153         else:
 154             return self.readall(size, num_retries=num_retries)
 155
 156     @_FileLikeObjectBase._before_close
 157     @retry_method
 158     def readlines(self, sizehint=float('inf'), num_retries=None):
 159         data = []
 160         data_size = 0
 161         for s in self.readall(num_retries=num_retries):
 162             data.append(s)
 163             data_size += len(s)
 164             if data_size >= sizehint:
 165                 break
 166         return ''.join(data).splitlines(True)
 167
 168     def size(self):
 169         raise NotImplementedError()
 170
 171     def read(self, size, num_retries=None):
 172         raise NotImplementedError()
 173
 174     def readfrom(self, start, size, num_retries=None):
 175         raise NotImplementedError()
 176
 177
 178 class StreamFileReader(ArvadosFileReaderBase):
 179     class _NameAttribute(str):
 180         # The Python file API provides a plain .name attribute.
 181         # Older SDK provided a name() method.
 182         # This class provides both, for maximum compatibility.
 183         def __call__(self):
 184             return self
 185
 186     def __init__(self, stream, segments, name):
 187         super(StreamFileReader, self).__init__(self._NameAttribute(name), 'rb', num_retries=stream.num_retries)
 188         self._stream = stream
 189         self.segments = segments
 190
 191     def stream_name(self):
 192         return self._stream.name()
 193
 194     def size(self):
 195         n = self.segments[-1]
 196         return n.range_start + n.range_size
 197
 198     @_FileLikeObjectBase._before_close
 199     @retry_method
 200     def read(self, size, num_retries=None):
 201         """Read up to 'size' bytes from the stream, starting at the current file position"""
 202         if size == 0:
 203             return ''
 204
 205         data = ''
 206         available_chunks = locators_and_ranges(self.segments, self._filepos, size)
 207         if available_chunks:
 208             lr = available_chunks[0]
 209             data = self._stream.readfrom(lr.locator+lr.segment_offset,
 210                                           lr.segment_size,
 211                                           num_retries=num_retries)
 212
 213         self._filepos += len(data)
 214         return data
 215
 216     @_FileLikeObjectBase._before_close
 217     @retry_method
 218     def readfrom(self, start, size, num_retries=None):
 219         """Read up to 'size' bytes from the stream, starting at 'start'"""
 220         if size == 0:
 221             return ''
 222
 223         data = []
 224         for lr in locators_and_ranges(self.segments, start, size):
 225             data.append(self._stream.readfrom(lr.locator+lr.segment_offset, lr.segment_size,
 226                                               num_retries=num_retries))
 227         return ''.join(data)
 228
 229     def as_manifest(self):
 230         segs = []
 231         for r in self.segments:
 232             segs.extend(self._stream.locators_and_ranges(r.locator, r.range_size))
 233         return " ".join(normalize_stream(".", {self.name: segs})) + "\n"
 234
 235
 236 def synchronized(orig_func):
 237     @functools.wraps(orig_func)
 238     def synchronized_wrapper(self, *args, **kwargs):
 239         with self.lock:
 240             return orig_func(self, *args, **kwargs)
 241     return synchronized_wrapper
 242
 243
 244 class StateChangeError(Exception):
 245     def __init__(self, message, state, nextstate):
 246         super(StateChangeError, self).__init__(message)
 247         self.state = state
 248         self.nextstate = nextstate
 249
 250 class _BufferBlock(object):
 251     """A stand-in for a Keep block that is in the process of being written.
 252
 253     Writers can append to it, get the size, and compute the Keep locator.
 254     There are three valid states:
 255
 256     WRITABLE
 257       Can append to block.
 258
 259     PENDING
 260       Block is in the process of being uploaded to Keep, append is an error.
 261
 262     COMMITTED
 263       The block has been written to Keep, its internal buffer has been
 264       released, fetching the block will fetch it via keep client (since we
 265       discarded the internal copy), and identifiers referring to the BufferBlock
 266       can be replaced with the block locator.
 267
 268     """
 269
 270     WRITABLE = 0
 271     PENDING = 1
 272     COMMITTED = 2
 273     ERROR = 3
 274
 275     def __init__(self, blockid, starting_capacity, owner):
 276         """
 277         :blockid:
 278           the identifier for this block
 279
 280         :starting_capacity:
 281           the initial buffer capacity
 282
 283         :owner:
 284           ArvadosFile that owns this block
 285
 286         """
 287         self.blockid = blockid
 288         self.buffer_block = bytearray(starting_capacity)
 289         self.buffer_view = memoryview(self.buffer_block)
 290         self.write_pointer = 0
 291         self._state = _BufferBlock.WRITABLE
 292         self._locator = None
 293         self.owner = owner
 294         self.lock = threading.Lock()
 295         self.wait_for_commit = threading.Event()
 296         self.error = None
 297
 298     @synchronized
 299     def append(self, data):
 300         """Append some data to the buffer.
 301
 302         Only valid if the block is in WRITABLE state.  Implements an expanding
 303         buffer, doubling capacity as needed to accomdate all the data.
 304
 305         """
 306         if self._state == _BufferBlock.WRITABLE:
 307             while (self.write_pointer+len(data)) > len(self.buffer_block):
 308                 new_buffer_block = bytearray(len(self.buffer_block) * 2)
 309                 new_buffer_block[0:self.write_pointer] = self.buffer_block[0:self.write_pointer]
 310                 self.buffer_block = new_buffer_block
 311                 self.buffer_view = memoryview(self.buffer_block)
 312             self.buffer_view[self.write_pointer:self.write_pointer+len(data)] = data
 313             self.write_pointer += len(data)
 314             self._locator = None
 315         else:
 316             raise AssertionError("Buffer block is not writable")
 317
 318     STATE_TRANSITIONS = frozenset([
 319             (WRITABLE, PENDING),
 320             (PENDING, COMMITTED),
 321             (PENDING, ERROR),
 322             (ERROR, PENDING)])
 323
 324     @synchronized
 325     def set_state(self, nextstate, val=None):
 326         if (self._state, nextstate) not in self.STATE_TRANSITIONS:
 327             raise StateChangeError("Invalid state change from %s to %s" % (self._state, nextstate), self._state, nextstate)
 328         self._state = nextstate
 329
 330         if self._state == _BufferBlock.PENDING:
 331             self.wait_for_commit.clear()
 332
 333         if self._state == _BufferBlock.COMMITTED:
 334             self._locator = val
 335             self.buffer_view = None
 336             self.buffer_block = None
 337             self.wait_for_commit.set()
 338
 339         if self._state == _BufferBlock.ERROR:
 340             self.error = val
 341             self.wait_for_commit.set()
 342
 343     @synchronized
 344     def state(self):
 345         return self._state
 346
 347     def size(self):
 348         """The amount of data written to the buffer."""
 349         return self.write_pointer
 350
 351     @synchronized
 352     def locator(self):
 353         """The Keep locator for this buffer's contents."""
 354         if self._locator is None:
 355             self._locator = "%s+%i" % (hashlib.md5(self.buffer_view[0:self.write_pointer]).hexdigest(), self.size())
 356         return self._locator
 357
 358     @synchronized
 359     def clone(self, new_blockid, owner):
 360         if self._state == _BufferBlock.COMMITTED:
 361             raise AssertionError("Cannot duplicate committed buffer block")
 362         bufferblock = _BufferBlock(new_blockid, self.size(), owner)
 363         bufferblock.append(self.buffer_view[0:self.size()])
 364         return bufferblock
 365
 366     @synchronized
 367     def clear(self):
 368         self.owner = None
 369         self.buffer_block = None
 370         self.buffer_view = None
 371
 372
 373 class NoopLock(object):
 374     def __enter__(self):
 375         return self
 376
 377     def __exit__(self, exc_type, exc_value, traceback):
 378         pass
 379
 380     def acquire(self, blocking=False):
 381         pass
 382
 383     def release(self):
 384         pass
 385
 386
 387 def must_be_writable(orig_func):
 388     @functools.wraps(orig_func)
 389     def must_be_writable_wrapper(self, *args, **kwargs):
 390         if not self.writable():
 391             raise IOError(errno.EROFS, "Collection is read-only.")
 392         return orig_func(self, *args, **kwargs)
 393     return must_be_writable_wrapper
 394
 395
 396 class _BlockManager(object):
 397     """BlockManager handles buffer blocks.
 398
 399     Also handles background block uploads, and background block prefetch for a
 400     Collection of ArvadosFiles.
 401
 402     """
 403
 404     DEFAULT_PUT_THREADS = 2
 405     DEFAULT_GET_THREADS = 2
 406
 407     def __init__(self, keep, copies=None, put_threads=None):
 408         """keep: KeepClient object to use"""
 409         self._keep = keep
 410         self._bufferblocks = collections.OrderedDict()
 411         self._put_queue = None
 412         self._put_threads = None
 413         self._prefetch_queue = None
 414         self._prefetch_threads = None
 415         self.lock = threading.Lock()
 416         self.prefetch_enabled = True
 417         if put_threads:
 418             self.num_put_threads = put_threads
 419         else:
 420             self.num_put_threads = _BlockManager.DEFAULT_PUT_THREADS
 421         self.num_get_threads = _BlockManager.DEFAULT_GET_THREADS
 422         self.copies = copies
 423         self._pending_write_size = 0
 424         self.threads_lock = threading.Lock()
 425
 426     @synchronized
 427     def alloc_bufferblock(self, blockid=None, starting_capacity=2**14, owner=None):
 428         """Allocate a new, empty bufferblock in WRITABLE state and return it.
 429
 430         :blockid:
 431           optional block identifier, otherwise one will be automatically assigned
 432
 433         :starting_capacity:
 434           optional capacity, otherwise will use default capacity
 435
 436         :owner:
 437           ArvadosFile that owns this block
 438
 439         """
 440         return self._alloc_bufferblock(blockid, starting_capacity, owner)
 441
 442     def _alloc_bufferblock(self, blockid=None, starting_capacity=2**14, owner=None):
 443         if blockid is None:
 444             blockid = "%s" % uuid.uuid4()
 445         bufferblock = _BufferBlock(blockid, starting_capacity=starting_capacity, owner=owner)
 446         self._bufferblocks[bufferblock.blockid] = bufferblock
 447         return bufferblock
 448
 449     @synchronized
 450     def dup_block(self, block, owner):
 451         """Create a new bufferblock initialized with the content of an existing bufferblock.
 452
 453         :block:
 454           the buffer block to copy.
 455
 456         :owner:
 457           ArvadosFile that owns the new block
 458
 459         """
 460         new_blockid = "bufferblock%i" % len(self._bufferblocks)
 461         bufferblock = block.clone(new_blockid, owner)
 462         self._bufferblocks[bufferblock.blockid] = bufferblock
 463         return bufferblock
 464
 465     @synchronized
 466     def is_bufferblock(self, locator):
 467         return locator in self._bufferblocks
 468
 469     def _commit_bufferblock_worker(self):
 470         """Background uploader thread."""
 471
 472         while True:
 473             try:
 474                 bufferblock = self._put_queue.get()
 475                 if bufferblock is None:
 476                     return
 477
 478                 if self.copies is None:
 479                     loc = self._keep.put(bufferblock.buffer_view[0:bufferblock.write_pointer].tobytes())
 480                 else:
 481                     loc = self._keep.put(bufferblock.buffer_view[0:bufferblock.write_pointer].tobytes(), copies=self.copies)
 482                 bufferblock.set_state(_BufferBlock.COMMITTED, loc)
 483
 484             except Exception as e:
 485                 bufferblock.set_state(_BufferBlock.ERROR, e)
 486             finally:
 487                 if self._put_queue is not None:
 488                     self._put_queue.task_done()
 489
 490     def start_put_threads(self):
 491         with self.threads_lock:
 492             if self._put_threads is None:
 493                 # Start uploader threads.
 494
 495                 # If we don't limit the Queue size, the upload queue can quickly
 496                 # grow to take up gigabytes of RAM if the writing process is
 497                 # generating data more quickly than it can be send to the Keep
 498                 # servers.
 499                 #
 500                 # With two upload threads and a queue size of 2, this means up to 4
 501                 # blocks pending.  If they are full 64 MiB blocks, that means up to
 502                 # 256 MiB of internal buffering, which is the same size as the
 503                 # default download block cache in KeepClient.
 504                 self._put_queue = Queue.Queue(maxsize=2)
 505
 506                 self._put_threads = []
 507                 for i in xrange(0, self.num_put_threads):
 508                     thread = threading.Thread(target=self._commit_bufferblock_worker)
 509                     self._put_threads.append(thread)
 510                     thread.daemon = True
 511                     thread.start()
 512
 513     def _block_prefetch_worker(self):
 514         """The background downloader thread."""
 515         while True:
 516             try:
 517                 b = self._prefetch_queue.get()
 518                 if b is None:
 519                     return
 520                 self._keep.get(b)
 521             except Exception:
 522                 _logger.exception("Exception doing block prefetch")
 523
 524     @synchronized
 525     def start_get_threads(self):
 526         if self._prefetch_threads is None:
 527             self._prefetch_queue = Queue.Queue()
 528             self._prefetch_threads = []
 529             for i in xrange(0, self.num_get_threads):
 530                 thread = threading.Thread(target=self._block_prefetch_worker)
 531                 self._prefetch_threads.append(thread)
 532                 thread.daemon = True
 533                 thread.start()
 534
 535
 536     @synchronized
 537     def stop_threads(self):
 538         """Shut down and wait for background upload and download threads to finish."""
 539
 540         if self._put_threads is not None:
 541             for t in self._put_threads:
 542                 self._put_queue.put(None)
 543             for t in self._put_threads:
 544                 t.join()
 545         self._put_threads = None
 546         self._put_queue = None
 547
 548         if self._prefetch_threads is not None:
 549             for t in self._prefetch_threads:
 550                 self._prefetch_queue.put(None)
 551             for t in self._prefetch_threads:
 552                 t.join()
 553         self._prefetch_threads = None
 554         self._prefetch_queue = None
 555
 556     def __enter__(self):
 557         return self
 558
 559     def __exit__(self, exc_type, exc_value, traceback):
 560         self.stop_threads()
 561
 562     @synchronized
 563     def repack_small_blocks(self, force=False, sync=False, closed_file_size=0):
 564         """Packs small blocks together before uploading"""
 565         self._pending_write_size += closed_file_size
 566
 567         # Check if there are enough small blocks for filling up one in full
 568         if force or (self._pending_write_size >= config.KEEP_BLOCK_SIZE):
 569
 570             # Search blocks ready for getting packed together before being committed to Keep.
 571             # A WRITABLE block always has an owner.
 572             # A WRITABLE block with its owner.closed() implies that it's
 573             # size is <= KEEP_BLOCK_SIZE/2.
 574             small_blocks = [b for b in self._bufferblocks.values() if b.state() == _BufferBlock.WRITABLE and b.owner.closed()]
 575
 576             if len(small_blocks) <= 1:
 577                 # Not enough small blocks for repacking
 578                 return
 579
 580             # Update the pending write size count with its true value, just in case
 581             # some small file was opened, written and closed several times.
 582             self._pending_write_size = sum([b.size() for b in small_blocks])
 583             if self._pending_write_size < config.KEEP_BLOCK_SIZE and not force:
 584                 return
 585
 586             new_bb = self._alloc_bufferblock()
 587             while len(small_blocks) > 0 and (new_bb.write_pointer + small_blocks[0].size()) <= config.KEEP_BLOCK_SIZE:
 588                 bb = small_blocks.pop(0)
 589                 arvfile = bb.owner
 590                 self._pending_write_size -= bb.size()
 591                 new_bb.append(bb.buffer_view[0:bb.write_pointer].tobytes())
 592                 arvfile.set_segments([Range(new_bb.blockid,
 593                                             0,
 594                                             bb.size(),
 595                                             new_bb.write_pointer - bb.size())])
 596                 self._delete_bufferblock(bb.blockid)
 597             self.commit_bufferblock(new_bb, sync=sync)
 598
 599     def commit_bufferblock(self, block, sync):
 600         """Initiate a background upload of a bufferblock.
 601
 602         :block:
 603           The block object to upload
 604
 605         :sync:
 606           If `sync` is True, upload the block synchronously.
 607           If `sync` is False, upload the block asynchronously.  This will
 608           return immediately unless the upload queue is at capacity, in
 609           which case it will wait on an upload queue slot.
 610
 611         """
 612         try:
 613             # Mark the block as PENDING so to disallow any more appends.
 614             block.set_state(_BufferBlock.PENDING)
 615         except StateChangeError as e:
 616             if e.state == _BufferBlock.PENDING:
 617                 if sync:
 618                     block.wait_for_commit.wait()
 619                 else:
 620                     return
 621             if block.state() == _BufferBlock.COMMITTED:
 622                 return
 623             elif block.state() == _BufferBlock.ERROR:
 624                 raise block.error
 625             else:
 626                 raise
 627
 628         if sync:
 629             try:
 630                 if self.copies is None:
 631                     loc = self._keep.put(block.buffer_view[0:block.write_pointer].tobytes())
 632                 else:
 633                     loc = self._keep.put(block.buffer_view[0:block.write_pointer].tobytes(), copies=self.copies)
 634                 block.set_state(_BufferBlock.COMMITTED, loc)
 635             except Exception as e:
 636                 block.set_state(_BufferBlock.ERROR, e)
 637                 raise
 638         else:
 639             self.start_put_threads()
 640             self._put_queue.put(block)
 641
 642     @synchronized
 643     def get_bufferblock(self, locator):
 644         return self._bufferblocks.get(locator)
 645
 646     @synchronized
 647     def delete_bufferblock(self, locator):
 648         self._delete_bufferblock(locator)
 649
 650     def _delete_bufferblock(self, locator):
 651         bb = self._bufferblocks[locator]
 652         bb.clear()
 653         del self._bufferblocks[locator]
 654
 655     def get_block_contents(self, locator, num_retries, cache_only=False):
 656         """Fetch a block.
 657
 658         First checks to see if the locator is a BufferBlock and return that, if
 659         not, passes the request through to KeepClient.get().
 660
 661         """
 662         with self.lock:
 663             if locator in self._bufferblocks:
 664                 bufferblock = self._bufferblocks[locator]
 665                 if bufferblock.state() != _BufferBlock.COMMITTED:
 666                     return bufferblock.buffer_view[0:bufferblock.write_pointer].tobytes()
 667                 else:
 668                     locator = bufferblock._locator
 669         if cache_only:
 670             return self._keep.get_from_cache(locator)
 671         else:
 672             return self._keep.get(locator, num_retries=num_retries)
 673
 674     def commit_all(self):
 675         """Commit all outstanding buffer blocks.
 676
 677         This is a synchronous call, and will not return until all buffer blocks
 678         are uploaded.  Raises KeepWriteError() if any blocks failed to upload.
 679
 680         """
 681         self.repack_small_blocks(force=True, sync=True)
 682
 683         with self.lock:
 684             items = self._bufferblocks.items()
 685
 686         for k,v in items:
 687             if v.state() != _BufferBlock.COMMITTED and v.owner:
 688                 v.owner.flush(sync=False)
 689
 690         with self.lock:
 691             if self._put_queue is not None:
 692                 self._put_queue.join()
 693
 694                 err = []
 695                 for k,v in items:
 696                     if v.state() == _BufferBlock.ERROR:
 697                         err.append((v.locator(), v.error))
 698                 if err:
 699                     raise KeepWriteError("Error writing some blocks", err, label="block")
 700
 701         for k,v in items:
 702             # flush again with sync=True to remove committed bufferblocks from
 703             # the segments.
 704             if v.owner:
 705                 v.owner.flush(sync=True)
 706
 707     def block_prefetch(self, locator):
 708         """Initiate a background download of a block.
 709
 710         This assumes that the underlying KeepClient implements a block cache,
 711         so repeated requests for the same block will not result in repeated
 712         downloads (unless the block is evicted from the cache.)  This method
 713         does not block.
 714
 715         """
 716
 717         if not self.prefetch_enabled:
 718             return
 719
 720         if self._keep.get_from_cache(locator) is not None:
 721             return
 722
 723         with self.lock:
 724             if locator in self._bufferblocks:
 725                 return
 726
 727         self.start_get_threads()
 728         self._prefetch_queue.put(locator)
 729
 730
 731 class ArvadosFile(object):
 732     """Represent a file in a Collection.
 733
 734     ArvadosFile manages the underlying representation of a file in Keep as a
 735     sequence of segments spanning a set of blocks, and implements random
 736     read/write access.
 737
 738     This object may be accessed from multiple threads.
 739
 740     """
 741
 742     def __init__(self, parent, name, stream=[], segments=[]):
 743         """
 744         ArvadosFile constructor.
 745
 746         :stream:
 747           a list of Range objects representing a block stream
 748
 749         :segments:
 750           a list of Range objects representing segments
 751         """
 752         self.parent = parent
 753         self.name = name
 754         self._writers = set()
 755         self._committed = False
 756         self._segments = []
 757         self.lock = parent.root_collection().lock
 758         for s in segments:
 759             self._add_segment(stream, s.locator, s.range_size)
 760         self._current_bblock = None
 761
 762     def writable(self):
 763         return self.parent.writable()
 764
 765     @synchronized
 766     def permission_expired(self, as_of_dt=None):
 767         """Returns True if any of the segment's locators is expired"""
 768         for r in self._segments:
 769             if KeepLocator(r.locator).permission_expired(as_of_dt):
 770                 return True
 771         return False
 772
 773     @synchronized
 774     def segments(self):
 775         return copy.copy(self._segments)
 776
 777     @synchronized
 778     def clone(self, new_parent, new_name):
 779         """Make a copy of this file."""
 780         cp = ArvadosFile(new_parent, new_name)
 781         cp.replace_contents(self)
 782         return cp
 783
 784     @must_be_writable
 785     @synchronized
 786     def replace_contents(self, other):
 787         """Replace segments of this file with segments from another `ArvadosFile` object."""
 788
 789         map_loc = {}
 790         self._segments = []
 791         for other_segment in other.segments():
 792             new_loc = other_segment.locator
 793             if other.parent._my_block_manager().is_bufferblock(other_segment.locator):
 794                 if other_segment.locator not in map_loc:
 795                     bufferblock = other.parent._my_block_manager().get_bufferblock(other_segment.locator)
 796                     if bufferblock.state() != _BufferBlock.WRITABLE:
 797                         map_loc[other_segment.locator] = bufferblock.locator()
 798                     else:
 799                         map_loc[other_segment.locator] = self.parent._my_block_manager().dup_block(bufferblock, self).blockid
 800                 new_loc = map_loc[other_segment.locator]
 801
 802             self._segments.append(Range(new_loc, other_segment.range_start, other_segment.range_size, other_segment.segment_offset))
 803
 804         self._committed = False
 805
 806     def __eq__(self, other):
 807         if other is self:
 808             return True
 809         if not isinstance(other, ArvadosFile):
 810             return False
 811
 812         othersegs = other.segments()
 813         with self.lock:
 814             if len(self._segments) != len(othersegs):
 815                 return False
 816             for i in xrange(0, len(othersegs)):
 817                 seg1 = self._segments[i]
 818                 seg2 = othersegs[i]
 819                 loc1 = seg1.locator
 820                 loc2 = seg2.locator
 821
 822                 if self.parent._my_block_manager().is_bufferblock(loc1):
 823                     loc1 = self.parent._my_block_manager().get_bufferblock(loc1).locator()
 824
 825                 if other.parent._my_block_manager().is_bufferblock(loc2):
 826                     loc2 = other.parent._my_block_manager().get_bufferblock(loc2).locator()
 827
 828                 if (KeepLocator(loc1).stripped() != KeepLocator(loc2).stripped() or
 829                     seg1.range_start != seg2.range_start or
 830                     seg1.range_size != seg2.range_size or
 831                     seg1.segment_offset != seg2.segment_offset):
 832                     return False
 833
 834         return True
 835
 836     def __ne__(self, other):
 837         return not self.__eq__(other)
 838
 839     @synchronized
 840     def set_segments(self, segs):
 841         self._segments = segs
 842
 843     @synchronized
 844     def set_committed(self):
 845         """Set committed flag to True"""
 846         self._committed = True
 847
 848     @synchronized
 849     def committed(self):
 850         """Get whether this is committed or not."""
 851         return self._committed
 852
 853     @synchronized
 854     def add_writer(self, writer):
 855         """Add an ArvadosFileWriter reference to the list of writers"""
 856         if isinstance(writer, ArvadosFileWriter):
 857             self._writers.add(writer)
 858
 859     @synchronized
 860     def remove_writer(self, writer, flush):
 861         """
 862         Called from ArvadosFileWriter.close(). Remove a writer reference from the list
 863         and do some block maintenance tasks.
 864         """
 865         self._writers.remove(writer)
 866
 867         if flush or self.size() > config.KEEP_BLOCK_SIZE / 2:
 868             # File writer closed, not small enough for repacking
 869             self.flush()
 870         elif self.closed():
 871             # All writers closed and size is adequate for repacking
 872             self.parent._my_block_manager().repack_small_blocks(closed_file_size=self.size())
 873
 874     def closed(self):
 875         """
 876         Get whether this is closed or not. When the writers list is empty, the file
 877         is supposed to be closed.
 878         """
 879         return len(self._writers) == 0
 880
 881     @must_be_writable
 882     @synchronized
 883     def truncate(self, size):
 884         """Shrink the size of the file.
 885
 886         If `size` is less than the size of the file, the file contents after
 887         `size` will be discarded.  If `size` is greater than the current size
 888         of the file, an IOError will be raised.
 889
 890         """
 891         if size < self.size():
 892             new_segs = []
 893             for r in self._segments:
 894                 range_end = r.range_start+r.range_size
 895                 if r.range_start >= size:
 896                     # segment is past the trucate size, all done
 897                     break
 898                 elif size < range_end:
 899                     nr = Range(r.locator, r.range_start, size - r.range_start, 0)
 900                     nr.segment_offset = r.segment_offset
 901                     new_segs.append(nr)
 902                     break
 903                 else:
 904                     new_segs.append(r)
 905
 906             self._segments = new_segs
 907             self._committed = False
 908         elif size > self.size():
 909             raise IOError(errno.EINVAL, "truncate() does not support extending the file size")
 910
 911     def readfrom(self, offset, size, num_retries, exact=False):
 912         """Read up to `size` bytes from the file starting at `offset`.
 913
 914         :exact:
 915          If False (default), return less data than requested if the read
 916          crosses a block boundary and the next block isn't cached.  If True,
 917          only return less data than requested when hitting EOF.
 918         """
 919
 920         with self.lock:
 921             if size == 0 or offset >= self.size():
 922                 return ''
 923             readsegs = locators_and_ranges(self._segments, offset, size)
 924             prefetch = locators_and_ranges(self._segments, offset + size, config.KEEP_BLOCK_SIZE, limit=32)
 925
 926         locs = set()
 927         data = []
 928         for lr in readsegs:
 929             block = self.parent._my_block_manager().get_block_contents(lr.locator, num_retries=num_retries, cache_only=(bool(data) and not exact))
 930             if block:
 931                 blockview = memoryview(block)
 932                 data.append(blockview[lr.segment_offset:lr.segment_offset+lr.segment_size].tobytes())
 933                 locs.add(lr.locator)
 934             else:
 935                 break
 936
 937         for lr in prefetch:
 938             if lr.locator not in locs:
 939                 self.parent._my_block_manager().block_prefetch(lr.locator)
 940                 locs.add(lr.locator)
 941
 942         return ''.join(data)
 943
 944     def _repack_writes(self, num_retries):
 945         """Test if the buffer block has more data than actual segments.
 946
 947         This happens when a buffered write over-writes a file range written in
 948         a previous buffered write.  Re-pack the buffer block for efficiency
 949         and to avoid leaking information.
 950
 951         """
 952         segs = self._segments
 953
 954         # Sum up the segments to get the total bytes of the file referencing
 955         # into the buffer block.
 956         bufferblock_segs = [s for s in segs if s.locator == self._current_bblock.blockid]
 957         write_total = sum([s.range_size for s in bufferblock_segs])
 958
 959         if write_total < self._current_bblock.size():
 960             # There is more data in the buffer block than is actually accounted for by segments, so
 961             # re-pack into a new buffer by copying over to a new buffer block.
 962             contents = self.parent._my_block_manager().get_block_contents(self._current_bblock.blockid, num_retries)
 963             new_bb = self.parent._my_block_manager().alloc_bufferblock(self._current_bblock.blockid, starting_capacity=write_total, owner=self)
 964             for t in bufferblock_segs:
 965                 new_bb.append(contents[t.segment_offset:t.segment_offset+t.range_size])
 966                 t.segment_offset = new_bb.size() - t.range_size
 967
 968             self._current_bblock = new_bb
 969
 970     @must_be_writable
 971     @synchronized
 972     def writeto(self, offset, data, num_retries):
 973         """Write `data` to the file starting at `offset`.
 974
 975         This will update existing bytes and/or extend the size of the file as
 976         necessary.
 977
 978         """
 979         if len(data) == 0:
 980             return
 981
 982         if offset > self.size():
 983             raise ArgumentError("Offset is past the end of the file")
 984
 985         if len(data) > config.KEEP_BLOCK_SIZE:
 986             # Chunk it up into smaller writes
 987             n = 0
 988             dataview = memoryview(data)
 989             while n < len(data):
 990                 self.writeto(offset+n, dataview[n:n + config.KEEP_BLOCK_SIZE].tobytes(), num_retries)
 991                 n += config.KEEP_BLOCK_SIZE
 992             return
 993
 994         self._committed = False
 995
 996         if self._current_bblock is None or self._current_bblock.state() != _BufferBlock.WRITABLE:
 997             self._current_bblock = self.parent._my_block_manager().alloc_bufferblock(owner=self)
 998
 999         if (self._current_bblock.size() + len(data)) > config.KEEP_BLOCK_SIZE:
1000             self._repack_writes(num_retries)
1001             if (self._current_bblock.size() + len(data)) > config.KEEP_BLOCK_SIZE:
1002                 self.parent._my_block_manager().commit_bufferblock(self._current_bblock, sync=False)
1003                 self._current_bblock = self.parent._my_block_manager().alloc_bufferblock(owner=self)
1004
1005         self._current_bblock.append(data)
1006
1007         replace_range(self._segments, offset, len(data), self._current_bblock.blockid, self._current_bblock.write_pointer - len(data))
1008
1009         self.parent.notify(WRITE, self.parent, self.name, (self, self))
1010
1011         return len(data)
1012
1013     @synchronized
1014     def flush(self, sync=True, num_retries=0):
1015         """Flush the current bufferblock to Keep.
1016
1017         :sync:
1018           If True, commit block synchronously, wait until buffer block has been written.
1019           If False, commit block asynchronously, return immediately after putting block into
1020           the keep put queue.
1021         """
1022         if self.committed():
1023             return
1024
1025         if self._current_bblock and self._current_bblock.state() != _BufferBlock.COMMITTED:
1026             if self._current_bblock.state() == _BufferBlock.WRITABLE:
1027                 self._repack_writes(num_retries)
1028             self.parent._my_block_manager().commit_bufferblock(self._current_bblock, sync=sync)
1029
1030         if sync:
1031             to_delete = set()
1032             for s in self._segments:
1033                 bb = self.parent._my_block_manager().get_bufferblock(s.locator)
1034                 if bb:
1035                     if bb.state() != _BufferBlock.COMMITTED:
1036                         self.parent._my_block_manager().commit_bufferblock(bb, sync=True)
1037                     to_delete.add(s.locator)
1038                     s.locator = bb.locator()
1039             for s in to_delete:
1040                self.parent._my_block_manager().delete_bufferblock(s)
1041
1042         self.parent.notify(MOD, self.parent, self.name, (self, self))
1043
1044     @must_be_writable
1045     @synchronized
1046     def add_segment(self, blocks, pos, size):
1047         """Add a segment to the end of the file.
1048
1049         `pos` and `offset` reference a section of the stream described by
1050         `blocks` (a list of Range objects)
1051
1052         """
1053         self._add_segment(blocks, pos, size)
1054
1055     def _add_segment(self, blocks, pos, size):
1056         """Internal implementation of add_segment."""
1057         self._committed = False
1058         for lr in locators_and_ranges(blocks, pos, size):
1059             last = self._segments[-1] if self._segments else Range(0, 0, 0, 0)
1060             r = Range(lr.locator, last.range_start+last.range_size, lr.segment_size, lr.segment_offset)
1061             self._segments.append(r)
1062
1063     @synchronized
1064     def size(self):
1065         """Get the file size."""
1066         if self._segments:
1067             n = self._segments[-1]
1068             return n.range_start + n.range_size
1069         else:
1070             return 0
1071
1072     @synchronized
1073     def manifest_text(self, stream_name=".", portable_locators=False, normalize=False):
1074         buf = ""
1075         filestream = []
1076         for segment in self.segments:
1077             loc = segment.locator
1078             if loc.startswith("bufferblock"):
1079                 loc = self._bufferblocks[loc].calculate_locator()
1080             if portable_locators:
1081                 loc = KeepLocator(loc).stripped()
1082             filestream.append(LocatorAndRange(loc, locator_block_size(loc),
1083                                  segment.segment_offset, segment.range_size))
1084         buf += ' '.join(normalize_stream(stream_name, {stream_name: filestream}))
1085         buf += "\n"
1086         return buf
1087
1088     @must_be_writable
1089     @synchronized
1090     def _reparent(self, newparent, newname):
1091         self._committed = False
1092         self.flush(sync=True)
1093         self.parent.remove(self.name)
1094         self.parent = newparent
1095         self.name = newname
1096         self.lock = self.parent.root_collection().lock
1097
1098
1099 class ArvadosFileReader(ArvadosFileReaderBase):
1100     """Wraps ArvadosFile in a file-like object supporting reading only.
1101
1102     Be aware that this class is NOT thread safe as there is no locking around
1103     updating file pointer.
1104
1105     """
1106
1107     def __init__(self, arvadosfile, num_retries=None):
1108         super(ArvadosFileReader, self).__init__(arvadosfile.name, "r", num_retries=num_retries)
1109         self.arvadosfile = arvadosfile
1110
1111     def size(self):
1112         return self.arvadosfile.size()
1113
1114     def stream_name(self):
1115         return self.arvadosfile.parent.stream_name()
1116
1117     @_FileLikeObjectBase._before_close
1118     @retry_method
1119     def read(self, size=None, num_retries=None):
1120         """Read up to `size` bytes from the file and return the result.
1121
1122         Starts at the current file position.  If `size` is None, read the
1123         entire remainder of the file.
1124         """
1125         if size is None:
1126             data = []
1127             rd = self.arvadosfile.readfrom(self._filepos, config.KEEP_BLOCK_SIZE, num_retries)
1128             while rd:
1129                 data.append(rd)
1130                 self._filepos += len(rd)
1131                 rd = self.arvadosfile.readfrom(self._filepos, config.KEEP_BLOCK_SIZE, num_retries)
1132             return ''.join(data)
1133         else:
1134             data = self.arvadosfile.readfrom(self._filepos, size, num_retries, exact=True)
1135             self._filepos += len(data)
1136             return data
1137
1138     @_FileLikeObjectBase._before_close
1139     @retry_method
1140     def readfrom(self, offset, size, num_retries=None):
1141         """Read up to `size` bytes from the stream, starting at the specified file offset.
1142
1143         This method does not change the file position.
1144         """
1145         return self.arvadosfile.readfrom(offset, size, num_retries)
1146
1147     def flush(self):
1148         pass
1149
1150
1151 class ArvadosFileWriter(ArvadosFileReader):
1152     """Wraps ArvadosFile in a file-like object supporting both reading and writing.
1153
1154     Be aware that this class is NOT thread safe as there is no locking around
1155     updating file pointer.
1156
1157     """
1158
1159     def __init__(self, arvadosfile, mode, num_retries=None):
1160         super(ArvadosFileWriter, self).__init__(arvadosfile, num_retries=num_retries)
1161         self.mode = mode
1162         self.arvadosfile.add_writer(self)
1163
1164     @_FileLikeObjectBase._before_close
1165     @retry_method
1166     def write(self, data, num_retries=None):
1167         if self.mode[0] == "a":
1168             self.arvadosfile.writeto(self.size(), data, num_retries)
1169         else:
1170             self.arvadosfile.writeto(self._filepos, data, num_retries)
1171             self._filepos += len(data)
1172         return len(data)
1173
1174     @_FileLikeObjectBase._before_close
1175     @retry_method
1176     def writelines(self, seq, num_retries=None):
1177         for s in seq:
1178             self.write(s, num_retries=num_retries)
1179
1180     @_FileLikeObjectBase._before_close
1181     def truncate(self, size=None):
1182         if size is None:
1183             size = self._filepos
1184         self.arvadosfile.truncate(size)
1185         if self._filepos > self.size():
1186             self._filepos = self.size()
1187
1188     @_FileLikeObjectBase._before_close
1189     def flush(self):
1190         self.arvadosfile.flush()
1191
1192     def close(self, flush=True):
1193         if not self.closed:
1194             self.arvadosfile.remove_writer(self, flush)
1195             super(ArvadosFileWriter, self).close()