proxystore.connectors.file

proxystore/connectors/file.py

File system connector implementation.

FileKey

Bases: NamedTuple

Key to objects in a file system directory.

Attributes:

• filename (str) –

  Unique object filename.

FileConnector

FileConnector(
    store_dir: str | PathLike[str],
    *,
    clear: bool = True,
    buffering: int = -1,
)

Connector to shared file system.

This connector writes objects to unique files within store_dir. Marker files are used to indicate that an object is finished being written to avoid race conditions.

Parameters:

• store_dir (str | PathLike[str]) –

  Path to directory to store data in. Note this directory will be deleted upon closing the store.

• clear (bool, default: True ) –

  Clear all objects on close() by removing store_dir.

• buffering (int, default: -1 ) –

  Buffering policy used with open().

Source code in proxystore/connectors/file.py

def __init__(
    self,
    store_dir: str | os.PathLike[str],
    *,
    clear: bool = True,
    buffering: int = -1,
) -> None:
    self.store_dir = os.path.abspath(store_dir)
    self.clear = clear
    self.buffering = buffering

    if not os.path.exists(self.store_dir):
        os.makedirs(self.store_dir, exist_ok=True)

close

close(*, clear: bool | None = None) -> bool

Close the connector and clean up.

Warning

This will delete the store_dir directory by default.

Note

A Store only clears its connector when closed if the store is the owner. Stores created implicitly, such as when resolving a proxy, never clear their connector.

Warning

This method should only be called at the end of the program when the connector will no longer be used, for example once all proxies have been resolved.

Parameters:

• clear (bool | None, default: None ) –

  Remove the store directory. Overrides the default value of clear provided when the FileConnector was instantiated.

Returns:

• bool –

  If the store directory was removed.

Source code in proxystore/connectors/file.py

def close(self, *, clear: bool | None = None) -> bool:
    """Close the connector and clean up.

    Warning:
        This will delete the `store_dir` directory by default.

    Note:
        A [`Store`][proxystore.store.base.Store] only clears its
        connector when closed if the store is the
        [`owner`][proxystore.store.base.Store.owner]. Stores created
        implicitly, such as when resolving a proxy, never clear their
        connector.

    Warning:
        This method should only be called at the end of the program
        when the connector will no longer be used, for example once all
        proxies have been resolved.

    Args:
        clear: Remove the store directory. Overrides the default
            value of `clear` provided when the
            [`FileConnector`][proxystore.connectors.file.FileConnector]
            was instantiated.

    Returns:
        If the store directory was removed.
    """
    clear = self.clear if clear is None else clear
    if clear and os.path.isdir(self.store_dir):
        shutil.rmtree(self.store_dir, ignore_errors=True)
    return clear

config

config() -> dict[str, Any]

Get the connector configuration.

The configuration contains all the information needed to reconstruct the connector object.

Source code in proxystore/connectors/file.py

def config(self) -> dict[str, Any]:
    """Get the connector configuration.

    The configuration contains all the information needed to reconstruct
    the connector object.
    """
    return {
        'store_dir': self.store_dir,
        'clear': self.clear,
        'buffering': self.buffering,
    }

from_config classmethod

from_config(config: dict[str, Any]) -> FileConnector

Create a new connector instance from a configuration.

Parameters:

• config (dict[str, Any]) –

  Configuration returned by .config().

Source code in proxystore/connectors/file.py

@classmethod
def from_config(cls, config: dict[str, Any]) -> FileConnector:
    """Create a new connector instance from a configuration.

    Args:
        config: Configuration returned by `#!python .config()`.
    """
    return cls(**init_kwargs(cls, config))

evict

evict(key: FileKey) -> None

Evict the object associated with the key.

Parameters:

• key (FileKey) –

  Key associated with object to evict.

Source code in proxystore/connectors/file.py

def evict(self, key: FileKey) -> None:
    """Evict the object associated with the key.

    Args:
        key: Key associated with object to evict.
    """
    path = os.path.join(self.store_dir, key.filename)
    marker = path + '.ready'
    # Remove the marker first so a concurrent get() does not see the
    # marker after the data file has been removed. Either file may
    # already have been removed by a concurrent evict().
    with contextlib.suppress(FileNotFoundError):
        os.remove(marker)
    with contextlib.suppress(FileNotFoundError):
        os.remove(path)

exists

exists(key: FileKey) -> bool

Check if an object associated with the key exists.

Parameters:

• key (FileKey) –

  Key potentially associated with stored object.

Returns:

• bool –

  If an object associated with the key exists.

Source code in proxystore/connectors/file.py

def exists(self, key: FileKey) -> bool:
    """Check if an object associated with the key exists.

    Args:
        key: Key potentially associated with stored object.

    Returns:
        If an object associated with the key exists.
    """
    path = os.path.join(self.store_dir, key.filename + '.ready')
    return os.path.exists(path)

get

get(key: FileKey) -> BytesLike | None

Get the serialized object associated with the key.

Parameters:

• key (FileKey) –

  Key associated with the object to retrieve.

Returns:

• BytesLike | None –

  Serialized object or None if the object does not exist.

Source code in proxystore/connectors/file.py

def get(self, key: FileKey) -> BytesLike | None:
    """Get the serialized object associated with the key.

    Args:
        key: Key associated with the object to retrieve.

    Returns:
        Serialized object or `None` if the object does not exist.
    """
    path = os.path.join(self.store_dir, key.filename)
    marker = path + '.ready'
    if os.path.exists(marker):
        try:
            with open(path, 'rb', buffering=self.buffering) as f:
                return f.read()
        except FileNotFoundError:
            # The object was evicted after the marker was checked.
            return None
    return None

get_batch

get_batch(
    keys: Sequence[FileKey],
) -> list[BytesLike | None]

Get a batch of serialized objects associated with the keys.

Parameters:

• keys (Sequence[FileKey]) –

  Sequence of keys associated with objects to retrieve.

Returns:

• list[BytesLike | None] –

  List with same order as keys with the serialized objects or

• list[BytesLike | None] –

  None if the corresponding key does not have an associated object.

Source code in proxystore/connectors/file.py

def get_batch(self, keys: Sequence[FileKey]) -> list[BytesLike | None]:
    """Get a batch of serialized objects associated with the keys.

    Args:
        keys: Sequence of keys associated with objects to retrieve.

    Returns:
        List with same order as `keys` with the serialized objects or
        `None` if the corresponding key does not have an associated object.
    """
    return [self.get(key) for key in keys]

new_key

new_key(obj: BytesLike | None = None) -> FileKey

Create a new key.

Parameters:

• obj (BytesLike | None, default: None ) –

  Optional object which the key will be associated with. Ignored by this implementation.

Returns:

• FileKey –

  Key which can be used to retrieve an object once set() has been called on the key.

Source code in proxystore/connectors/file.py

def new_key(self, obj: BytesLike | None = None) -> FileKey:
    """Create a new key.

    Args:
        obj: Optional object which the key will be associated with.
            Ignored by this implementation.

    Returns:
        Key which can be used to retrieve an object once \
        [`set()`][proxystore.connectors.file.FileConnector.set] \
        has been called on the key.
    """
    return FileKey(filename=str(uuid.uuid4()))

put

put(obj: BytesLike) -> FileKey

Put a serialized object in the store.

Parameters:

• obj (BytesLike) –

  Serialized object to put in the store.

Returns:

• FileKey –

  Key which can be used to retrieve the object.

Source code in proxystore/connectors/file.py

def put(self, obj: BytesLike) -> FileKey:
    """Put a serialized object in the store.

    Args:
        obj: Serialized object to put in the store.

    Returns:
        Key which can be used to retrieve the object.
    """
    key = FileKey(filename=str(uuid.uuid4()))
    self.set(key, obj)
    return key

put_batch

put_batch(objs: Sequence[BytesLike]) -> list[FileKey]

Put a batch of serialized objects in the store.

Parameters:

• objs (Sequence[BytesLike]) –

  Sequence of serialized objects to put in the store.

Returns:

• list[FileKey] –

  List of keys with the same order as objs which can be used to

• list[FileKey] –

  retrieve the objects.

Source code in proxystore/connectors/file.py

def put_batch(self, objs: Sequence[BytesLike]) -> list[FileKey]:
    """Put a batch of serialized objects in the store.

    Args:
        objs: Sequence of serialized objects to put in the store.

    Returns:
        List of keys with the same order as `objs` which can be used to
        retrieve the objects.
    """
    return [self.put(obj) for obj in objs]

set

set(key: FileKey, obj: BytesLike) -> None

Set the object associated with a key.

Note

The Connector provides write-once, read-many semantics. Thus, set() should only be called once per key, otherwise unexpected behavior can occur.

Parameters:

• key (FileKey) –

  Key that the object will be associated with.

• obj (BytesLike) –

  Object to associate with the key.

Source code in proxystore/connectors/file.py

def set(self, key: FileKey, obj: BytesLike) -> None:
    """Set the object associated with a key.

    Note:
        The [`Connector`][proxystore.connectors.protocols.Connector]
        provides write-once, read-many semantics. Thus,
        [`set()`][proxystore.connectors.file.FileConnector.set]
        should only be called once per key, otherwise unexpected behavior
        can occur.

    Args:
        key: Key that the object will be associated with.
        obj: Object to associate with the key.
    """
    path = os.path.join(self.store_dir, key.filename)
    with open(path, 'wb', buffering=self.buffering) as f:
        f.write(obj)
    marker = path + '.ready'
    open(marker, 'wb').close()