X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/682dd5b6cc23a455766a7651e3e841257660b31c..8c5f2973a5c5f042d1d12aef1c470b37519fd416:/sdk/python/arvados/collection.py diff --git a/sdk/python/arvados/collection.py b/sdk/python/arvados/collection.py index 89cbda937c..f26d3a3d27 100644 --- a/sdk/python/arvados/collection.py +++ b/sdk/python/arvados/collection.py @@ -307,7 +307,8 @@ class CollectionWriter(CollectionBase): def set_current_stream_name(self, newstreamname): if re.search(r'[\t\n]', newstreamname): raise errors.AssertionError( - "Manifest stream names cannot contain whitespace") + "Manifest stream names cannot contain whitespace: '%s'" % + (newstreamname)) self._current_stream_name = '.' if newstreamname=='' else newstreamname def current_stream_name(self): @@ -474,6 +475,7 @@ class ResumableCollectionWriter(CollectionWriter): ADD = "add" DEL = "del" MOD = "mod" +TOK = "tok" FILE = "file" COLLECTION = "collection" @@ -487,7 +489,7 @@ class RichCollectionBase(CollectionBase): def __init__(self, parent=None): self.parent = parent - self._modified = True + self._committed = False self._callback = None self._items = {} @@ -542,7 +544,7 @@ class RichCollectionBase(CollectionBase): else: item = ArvadosFile(self, pathcomponents[0]) self._items[pathcomponents[0]] = item - self._modified = True + self.set_committed(False) self.notify(ADD, self, pathcomponents[0], item) return item else: @@ -550,12 +552,12 @@ class RichCollectionBase(CollectionBase): # create new collection item = Subcollection(self, pathcomponents[0]) self._items[pathcomponents[0]] = item - self._modified = True + self.set_committed(False) self.notify(ADD, self, pathcomponents[0], item) if isinstance(item, RichCollectionBase): return item.find_or_create(pathcomponents[1], create_type) else: - raise IOError(errno.ENOTDIR, "Interior path components must be subcollection") + raise IOError(errno.ENOTDIR, "Not a directory", pathcomponents[0]) else: return self @@ -563,16 +565,23 @@ class RichCollectionBase(CollectionBase): def find(self, path): """Recursively search the specified file path. - May return either a Collection or ArvadosFile. Return None if not + May return either a Collection or ArvadosFile. Return None if not found. + If path is invalid (ex: starts with '/'), an IOError exception will be + raised. """ if not path: - raise errors.ArgumentError("Parameter 'path' must not be empty.") + raise errors.ArgumentError("Parameter 'path' is empty.") pathcomponents = path.split("/", 1) + if pathcomponents[0] == '': + raise IOError(errno.ENOTDIR, "Not a directory", pathcomponents[0]) + item = self._items.get(pathcomponents[0]) - if len(pathcomponents) == 1: + if item is None: + return None + elif len(pathcomponents) == 1: return item else: if isinstance(item, RichCollectionBase): @@ -581,15 +590,20 @@ class RichCollectionBase(CollectionBase): else: return item else: - raise IOError(errno.ENOTDIR, "Interior path components must be subcollection") + raise IOError(errno.ENOTDIR, "Not a directory", pathcomponents[0]) + @synchronized def mkdirs(self, path): """Recursive subcollection create. - Like `os.mkdirs()`. Will create intermediate subcollections needed to - contain the leaf subcollection path. + Like `os.makedirs()`. Will create intermediate subcollections needed + to contain the leaf subcollection path. """ + + if self.find(path) != None: + raise IOError(errno.EEXIST, "Directory or file exists", path) + return self.find_or_create(path, COLLECTION) def open(self, path, mode="r"): @@ -624,9 +638,9 @@ class RichCollectionBase(CollectionBase): arvfile = self.find(path) if arvfile is None: - raise IOError(errno.ENOENT, "File not found") + raise IOError(errno.ENOENT, "File not found", path) if not isinstance(arvfile, ArvadosFile): - raise IOError(errno.EISDIR, "Path must refer to a file.") + raise IOError(errno.EISDIR, "Is a directory", path) if mode[0] == "w": arvfile.truncate(0) @@ -634,26 +648,37 @@ class RichCollectionBase(CollectionBase): name = os.path.basename(path) if mode == "r": - return ArvadosFileReader(arvfile, mode, num_retries=self.num_retries) + return ArvadosFileReader(arvfile, num_retries=self.num_retries) else: return ArvadosFileWriter(arvfile, mode, num_retries=self.num_retries) - @synchronized def modified(self): - """Test if the collection (or any subcollection or file) has been modified.""" - if self._modified: - return True - for k,v in self._items.items(): - if v.modified(): - return True - return False + """Determine if the collection has been modified since last commited.""" + return not self.committed() @synchronized - def set_unmodified(self): - """Recursively clear modified flag.""" - self._modified = False - for k,v in self._items.items(): - v.set_unmodified() + def committed(self): + """Determine if the collection has been committed to the API server.""" + return self._committed + + @synchronized + def set_committed(self, value=True): + """Recursively set committed flag. + + If value is True, set committed to be True for this and all children. + + If value is False, set committed to be False for this and all parents. + """ + if value == self._committed: + return + if value: + for k,v in self._items.items(): + v.set_committed(True) + self._committed = True + else: + self._committed = False + if self.parent is not None: + self.parent.set_committed(False) @synchronized def __iter__(self): @@ -684,7 +709,7 @@ class RichCollectionBase(CollectionBase): def __delitem__(self, p): """Delete an item by name which is directly contained by this collection.""" del self._items[p] - self._modified = True + self.set_committed(False) self.notify(DEL, self, p, None) @synchronized @@ -716,18 +741,18 @@ class RichCollectionBase(CollectionBase): """ if not path: - raise errors.ArgumentError("Parameter 'path' must not be empty.") + raise errors.ArgumentError("Parameter 'path' is empty.") pathcomponents = path.split("/", 1) item = self._items.get(pathcomponents[0]) if item is None: - raise IOError(errno.ENOENT, "File not found") + raise IOError(errno.ENOENT, "File not found", path) if len(pathcomponents) == 1: if isinstance(self._items[pathcomponents[0]], RichCollectionBase) and len(self._items[pathcomponents[0]]) > 0 and not recursive: - raise IOError(errno.ENOTEMPTY, "Subcollection not empty") + raise IOError(errno.ENOTEMPTY, "Directory not empty", path) deleteditem = self._items[pathcomponents[0]] del self._items[pathcomponents[0]] - self._modified = True + self.set_committed(False) self.notify(DEL, self, pathcomponents[0], deleteditem) else: item.remove(pathcomponents[1]) @@ -762,7 +787,7 @@ class RichCollectionBase(CollectionBase): """ if target_name in self and not overwrite: - raise IOError(errno.EEXIST, "File already exists") + raise IOError(errno.EEXIST, "File already exists", target_name) modified_from = None if target_name in self: @@ -770,13 +795,13 @@ class RichCollectionBase(CollectionBase): # Actually make the move or copy. if reparent: - source_obj.reparent(self, target_name) + source_obj._reparent(self, target_name) item = source_obj else: item = source_obj.clone(self, target_name) self._items[target_name] = item - self._modified = True + self.set_committed(False) if modified_from: self.notify(MOD, self, target_name, (modified_from, item)) @@ -791,7 +816,7 @@ class RichCollectionBase(CollectionBase): if isinstance(source, basestring): source_obj = source_collection.find(source) if source_obj is None: - raise IOError(errno.ENOENT, "File not found") + raise IOError(errno.ENOENT, "File not found", source) sourcecomponents = source.split("/") else: source_obj = source @@ -815,9 +840,9 @@ class RichCollectionBase(CollectionBase): target_dir = self if target_dir is None: - raise IOError(errno.ENOENT, "Target directory not found.") + raise IOError(errno.ENOENT, "Target directory not found", target_name) - if target_name in target_dir and isinstance(self[target_name], RichCollectionBase) and sourcecomponents: + if target_name in target_dir and isinstance(target_dir[target_name], RichCollectionBase) and sourcecomponents: target_dir = target_dir[target_name] target_name = sourcecomponents[-1] @@ -870,7 +895,7 @@ class RichCollectionBase(CollectionBase): source_obj, target_dir, target_name = self._get_src_target(source, target_path, source_collection, False) if not source_obj.writable(): - raise IOError(errno.EROFS, "Source collection must be writable.") + raise IOError(errno.EROFS, "Source collection is read only", source) target_dir.add(source_obj, target_name, overwrite, True) def portable_manifest_text(self, stream_name="."): @@ -885,7 +910,9 @@ class RichCollectionBase(CollectionBase): """ return self._get_manifest_text(stream_name, True, True) - def manifest_text(self, stream_name=".", strip=False, normalize=False): + @synchronized + def manifest_text(self, stream_name=".", strip=False, normalize=False, + only_committed=False): """Get the manifest text for this collection, sub collections and files. This method will flush outstanding blocks to Keep. By default, it will @@ -904,13 +931,18 @@ class RichCollectionBase(CollectionBase): is not modified, return the original manifest text even if it is not in normalized form. + :only_committed: + If True, don't commit pending blocks. + """ - self._my_block_manager().commit_all() - return self._get_manifest_text(stream_name, strip, normalize) + if not only_committed: + self._my_block_manager().commit_all() + return self._get_manifest_text(stream_name, strip, normalize, + only_committed=only_committed) @synchronized - def _get_manifest_text(self, stream_name, strip, normalize): + def _get_manifest_text(self, stream_name, strip, normalize, only_committed=False): """Get the manifest text for this collection, sub collections and files. :stream_name: @@ -926,9 +958,12 @@ class RichCollectionBase(CollectionBase): is not modified, return the original manifest text even if it is not in normalized form. + :only_committed: + If True, only include blocks that were already committed to Keep. + """ - if self.modified() or self._manifest_text is None or normalize: + if not self.committed() or self._manifest_text is None or normalize: stream = {} buf = [] sorted_keys = sorted(self.keys()) @@ -939,6 +974,8 @@ class RichCollectionBase(CollectionBase): for segment in arvfile.segments(): loc = segment.locator if arvfile.parent._my_block_manager().is_bufferblock(loc): + if only_committed: + continue loc = arvfile.parent._my_block_manager().get_bufferblock(loc).locator() if strip: loc = KeepLocator(loc).stripped() @@ -948,7 +985,7 @@ class RichCollectionBase(CollectionBase): if stream: buf.append(" ".join(normalize_stream(stream_name, stream)) + "\n") for dirname in [s for s in sorted_keys if isinstance(self[s], RichCollectionBase)]: - buf.append(self[dirname].manifest_text(stream_name=os.path.join(stream_name, dirname), strip=strip, normalize=True)) + buf.append(self[dirname].manifest_text(stream_name=os.path.join(stream_name, dirname), strip=strip, normalize=True, only_committed=only_committed)) return "".join(buf) else: if strip: @@ -975,6 +1012,8 @@ class RichCollectionBase(CollectionBase): changes.extend(self[k].diff(end_collection[k], os.path.join(prefix, k), holding_collection)) elif end_collection[k] != self[k]: changes.append((MOD, os.path.join(prefix, k), self[k].clone(holding_collection, ""), end_collection[k].clone(holding_collection, ""))) + else: + changes.append((TOK, os.path.join(prefix, k), self[k].clone(holding_collection, ""), end_collection[k].clone(holding_collection, ""))) else: changes.append((ADD, os.path.join(prefix, k), end_collection[k].clone(holding_collection, ""))) return changes @@ -988,12 +1027,14 @@ class RichCollectionBase(CollectionBase): alternate path indicating the conflict. """ + if changes: + self.set_committed(False) for change in changes: event_type = change[0] path = change[1] initial = change[2] local = self.find(path) - conflictpath = "%s~conflict-%s~" % (path, time.strftime("%Y-%m-%d-%H:%M:%S", + conflictpath = "%s~%s~conflict~" % (path, time.strftime("%Y%m%d-%H%M%S", time.gmtime())) if event_type == ADD: if local is None: @@ -1003,7 +1044,7 @@ class RichCollectionBase(CollectionBase): # There is already local file and it is different: # save change to conflict file. self.copy(initial, conflictpath) - elif event_type == MOD: + elif event_type == MOD or event_type == TOK: final = change[3] if local == initial: # Local matches the "initial" item so it has not @@ -1028,8 +1069,13 @@ class RichCollectionBase(CollectionBase): def portable_data_hash(self): """Get the portable data hash for this collection's manifest.""" - stripped = self.portable_manifest_text() - return hashlib.md5(stripped).hexdigest() + '+' + str(len(stripped)) + if self._manifest_locator and self.committed(): + # If the collection is already saved on the API server, and it's committed + # then return API server's PDH response. + return self._portable_data_hash + else: + stripped = self.portable_manifest_text() + return hashlib.md5(stripped).hexdigest() + '+' + str(len(stripped)) @synchronized def subscribe(self, callback): @@ -1067,6 +1113,12 @@ class RichCollectionBase(CollectionBase): def __ne__(self, other): return not self.__eq__(other) + @synchronized + def flush(self): + """Flush bufferblocks to Keep.""" + for e in self.values(): + e.flush() + class Collection(RichCollectionBase): """Represents the root of an Arvados Collection. @@ -1113,7 +1165,9 @@ class Collection(RichCollectionBase): num_retries=None, parent=None, apiconfig=None, - block_manager=None): + block_manager=None, + replication_desired=None, + put_threads=None): """Collection constructor. :manifest_locator_or_text: @@ -1121,24 +1175,36 @@ class Collection(RichCollectionBase): a manifest, raw manifest text, or None (to create an empty collection). :parent: the parent Collection, may be None. + :apiconfig: A dict containing keys for ARVADOS_API_HOST and ARVADOS_API_TOKEN. Prefer this over supplying your own api_client and keep_client (except in testing). Will use default config settings if not specified. + :api_client: The API client object to use for requests. If not specified, create one using `apiconfig`. + :keep_client: the Keep client to use for requests. If not specified, create one using `apiconfig`. + :num_retries: the number of retries for API and Keep requests. + :block_manager: the block manager to use. If not specified, create one. + :replication_desired: + How many copies should Arvados maintain. If None, API server default + configuration applies. If not None, this value will also be used + for determining the number of block copies being written. + """ super(Collection, self).__init__(parent) self._api_client = api_client self._keep_client = keep_client self._block_manager = block_manager + self.replication_desired = replication_desired + self.put_threads = put_threads if apiconfig: self._config = apiconfig @@ -1148,7 +1214,9 @@ class Collection(RichCollectionBase): self.num_retries = num_retries if num_retries is not None else 0 self._manifest_locator = None self._manifest_text = None + self._portable_data_hash = None self._api_response = None + self._past_versions = set() self.lock = threading.RLock() self.events = None @@ -1162,7 +1230,7 @@ class Collection(RichCollectionBase): self._manifest_text = manifest_locator_or_text else: raise errors.ArgumentError( - "Argument to CollectionReader must be a manifest or a collection UUID") + "Argument to CollectionReader is not a manifest or a collection UUID") try: self._populate() @@ -1178,6 +1246,10 @@ class Collection(RichCollectionBase): def writable(self): return True + @synchronized + def known_past_version(self, modified_at_and_portable_data_hash): + return modified_at_and_portable_data_hash in self._past_versions + @synchronized @retry_method def update(self, other=None, num_retries=None): @@ -1187,15 +1259,26 @@ class Collection(RichCollectionBase): if self._manifest_locator is None: raise errors.ArgumentError("`other` is None but collection does not have a manifest_locator uuid") response = self._my_api().collections().get(uuid=self._manifest_locator).execute(num_retries=num_retries) + if (self.known_past_version((response.get("modified_at"), response.get("portable_data_hash"))) and + response.get("portable_data_hash") != self.portable_data_hash()): + # The record on the server is different from our current one, but we've seen it before, + # so ignore it because it's already been merged. + # However, if it's the same as our current record, proceed with the update, because we want to update + # our tokens. + return + else: + self._past_versions.add((response.get("modified_at"), response.get("portable_data_hash"))) other = CollectionReader(response["manifest_text"]) baseline = CollectionReader(self._manifest_text) self.apply(baseline.diff(other)) + self._manifest_text = self.manifest_text() @synchronized def _my_api(self): if self._api_client is None: self._api_client = ThreadSafeApiCache(self._config) - self._keep_client = self._api_client.keep + if self._keep_client is None: + self._keep_client = self._api_client.keep return self._api_client @synchronized @@ -1210,9 +1293,16 @@ class Collection(RichCollectionBase): @synchronized def _my_block_manager(self): if self._block_manager is None: - self._block_manager = _BlockManager(self._my_keep()) + copies = (self.replication_desired or + self._my_api()._rootDesc.get('defaultCollectionReplication', + 2)) + self._block_manager = _BlockManager(self._my_keep(), copies=copies, put_threads=self.put_threads) return self._block_manager + def _remember_api_response(self, response): + self._api_response = response + self._past_versions.add((response.get("modified_at"), response.get("portable_data_hash"))) + def _populate_from_api_server(self): # As in KeepClient itself, we must wait until the last # possible moment to instantiate an API client, in order to @@ -1222,10 +1312,15 @@ class Collection(RichCollectionBase): # clause, just like any other Collection lookup # failure. Return an exception, or None if successful. try: - self._api_response = self._my_api().collections().get( + self._remember_api_response(self._my_api().collections().get( uuid=self._manifest_locator).execute( - num_retries=self.num_retries) + num_retries=self.num_retries)) self._manifest_text = self._api_response['manifest_text'] + self._portable_data_hash = self._api_response['portable_data_hash'] + # If not overriden via kwargs, we should try to load the + # replication_desired from the API server + if self.replication_desired is None: + self.replication_desired = self._api_response.get('replication_desired', None) return None except Exception as e: return e @@ -1286,6 +1381,9 @@ class Collection(RichCollectionBase): if exc_type is None: if self.writable() and self._has_collection_uuid(): self.save() + self.stop_threads() + + def stop_threads(self): if self._block_manager is not None: self._block_manager.stop_threads() @@ -1368,9 +1466,9 @@ class Collection(RichCollectionBase): Retry count on API calls (if None, use the collection default) """ - if self.modified(): + if not self.committed(): if not self._has_collection_uuid(): - raise AssertionError("Collection manifest_locator must be a collection uuid. Use save_new() for new collections.") + raise AssertionError("Collection manifest_locator is not a collection uuid. Use save_new() for new collections.") self._my_block_manager().commit_all() @@ -1378,13 +1476,14 @@ class Collection(RichCollectionBase): self.update() text = self.manifest_text(strip=False) - self._api_response = self._my_api().collections().update( + self._remember_api_response(self._my_api().collections().update( uuid=self._manifest_locator, body={'manifest_text': text} ).execute( - num_retries=num_retries) + num_retries=num_retries)) self._manifest_text = self._api_response["manifest_text"] - self.set_unmodified() + self._portable_data_hash = self._api_response["portable_data_hash"] + self.set_committed(True) return self._manifest_text @@ -1429,21 +1528,23 @@ class Collection(RichCollectionBase): if create_collection_record: if name is None: - name = "Collection created %s" % (time.strftime("%Y-%m-%d %H:%M:%S %Z", time.localtime())) + name = "New collection" ensure_unique_name = True body = {"manifest_text": text, - "name": name} + "name": name, + "replication_desired": self.replication_desired} if owner_uuid: body["owner_uuid"] = owner_uuid - self._api_response = self._my_api().collections().create(ensure_unique_name=ensure_unique_name, body=body).execute(num_retries=num_retries) + self._remember_api_response(self._my_api().collections().create(ensure_unique_name=ensure_unique_name, body=body).execute(num_retries=num_retries)) text = self._api_response["manifest_text"] self._manifest_locator = self._api_response["uuid"] + self._portable_data_hash = self._api_response["portable_data_hash"] self._manifest_text = text - self.set_unmodified() + self.set_committed(True) return text @@ -1476,7 +1577,7 @@ class Collection(RichCollectionBase): segments = [] streamoffset = 0L state = BLOCKS - self.mkdirs(stream_name) + self.find_or_create(stream_name, COLLECTION) continue if state == BLOCKS: @@ -1502,13 +1603,13 @@ class Collection(RichCollectionBase): raise errors.SyntaxError("File %s conflicts with stream of the same name.", filepath) else: # error! - raise errors.SyntaxError("Invalid manifest format") + raise errors.SyntaxError("Invalid manifest format, expected file segment but did not match format: '%s'" % tok) if sep == "\n": stream_name = None state = STREAM_NAME - self.set_unmodified() + self.set_committed(True) @synchronized def notify(self, event, collection, name, item): @@ -1520,7 +1621,7 @@ class Subcollection(RichCollectionBase): """This is a subdirectory within a collection that doesn't have its own API server record. - It falls under the umbrella of the root collection. + Subcollection locking falls under the umbrella lock of its root collection. """ @@ -1555,6 +1656,16 @@ class Subcollection(RichCollectionBase): c._clonefrom(self) return c + @must_be_writable + @synchronized + def _reparent(self, newparent, newname): + self.set_committed(False) + self.flush() + self.parent.remove(self.name, recursive=True) + self.parent = newparent + self.name = newname + self.lock = self.parent.root_collection().lock + class CollectionReader(Collection): """A read-only collection object.