dragon.workflows.batch.batch.Function

class Function[source]

Bases: Task

__init__(batch, target: Callable , args: tuple = (), kwargs: dict = {}, reads: list | None = None, writes: list | None = None, name: str | None = None, timeout: float = 1000000000.0, stdout: str | Path | None = None, stderr: str | Path | None = None) → None [source]

Creates a new function task. Arguments for the function that are of type Task will create a dependency for this task on the output of the task specified by the argument. Further, the output of the specified task will be passed in place of the Task argument when the function executes.

Parameters:

  • batch – The batch in which this function task will execute.

  • func – The function to associate with the object.

  • *args –

    The arguments for the function.

  • reads (Optional[list ]) – A list of Read objects created by calling Batch.read().

  • writes (Optional[list ]) – A list of Write objects created by calling Batch.write().

  • name (Optional[str ]) – A human-readable name for the task.

Returns:

Returns None.

Return type:

None

Methods

__init__(batch, target[, args, kwargs, ...])

Creates a new function task.

cancel([timeout])

Best-effort cancellation request for this task.

get([block, timeout])

Wait for this Task to complete and return its result.

get_manager_idx([block, timeout])

get_stderr([block, timeout, encoding])

Read this task's stderr log.

get_stdout([block, timeout, encoding])

Read this task's stdout log.

log_paths()

Return the resolved stdout/stderr log file paths for this task.

Attributes

stderr_path

Return the resolved stderr log path for this task, if any.

stdout_path

Return the resolved stdout log path for this task, if any.

uid

Provides the unique ID for this task.

weight

Provides the HEFTY weight for this task.
__init__(batch, target: Callable , args: tuple = (), kwargs: dict = {}, reads: list | None = None, writes: list | None = None, name: str | None = None, timeout: float = 1000000000.0, stdout: str | Path | None = None, stderr: str | Path | None = None) → None [source]

Creates a new function task. Arguments for the function that are of type Task will create a dependency for this task on the output of the task specified by the argument. Further, the output of the specified task will be passed in place of the Task argument when the function executes.

Parameters:

  • batch – The batch in which this function task will execute.

  • func – The function to associate with the object.

  • *args –

    The arguments for the function.

  • reads (Optional[list ]) – A list of Read objects created by calling Batch.read().

  • writes (Optional[list ]) – A list of Write objects created by calling Batch.write().

  • name (Optional[str ]) – A human-readable name for the task.

Returns:

Returns None.

Return type:

None

cancel(timeout: float = 1000000000.0) → bool [source]

Best-effort cancellation request for this task.

Returns True when the scheduler accepted the cancellation and the task will complete with TaskCancelledError. Returns False when the task cannot be cancelled or already completed.

get(block: bool = True, timeout: float = 1000000000.0) → None [source]

Wait for this Task to complete and return its result.

get() never prints task stdout or stderr. When the owning Batch was created with task_logs=True, task stdout and stderr are redirected to files; use Task.stdout_path(), Task.stderr_path(), Task.get_stdout(), Task.get_stderr(), or Task.log_paths() to inspect the resolved task logs. When task logging is disabled (the default), task stdout and stderr are not captured to files unless an explicit stdout/stderr path was given; otherwise they are forwarded to the client console like any other Dragon process. Task results remain in Batch’s internal results DDict until get() fetches them or Batch.clear_results() discards them. Once fetched here, the result is cached on this client-side task handle for subsequent calls. In unmanaged mode, fetch any result you need before the last client calls Batch.join(), because that shutdown destroys the shared results DDict.

Parameters:

  • block (bool ) – If True (the default), block until the result is available or timeout is exceeded. If False, return immediately if the result is available, otherwise raise TaskNotReadyError.

  • timeout (float ) – The timeout for waiting. Defaults to 1e9. Ignored when block is False.

Raises:

TimeoutError – If the specified timeout is exceeded.

:raises TaskNotReadyError: If block is False and the result is not yet available.

Returns:

Returns the result of the task.

Return type:

Any

get_manager_idx(block: bool = True, timeout: float = 1000000000.0) → int [source]
get_stderr(block: bool = True, timeout: float = 1000000000.0, encoding: str = 'utf-8') → str | None [source]

Read this task’s stderr log.

By default, waits for the task to complete before reading the final log file. When block is False, raises TaskNotReadyError if the task has not completed yet.

get_stdout(block: bool = True, timeout: float = 1000000000.0, encoding: str = 'utf-8') → str | None [source]

Read this task’s stdout log.

By default, waits for the task to complete before reading the final log file. When block is False, raises TaskNotReadyError if the task has not completed yet.

log_paths() → dict [str , str | None ][source]

Return the resolved stdout/stderr log file paths for this task.

The returned dictionary contains stdout and stderr keys whose values are the shared-filesystem paths configured or generated for this task. A value is None when no path was configured, which happens for tasks submitted without an explicit stdout/stderr on a client created with task_logs=False (the default), since their output is forwarded to the client console instead of a file.

property stderr_path: str | None

Return the resolved stderr log path for this task, if any.

property stdout_path: str | None

Return the resolved stdout log path for this task, if any.

property uid

Provides the unique ID for this task.

property weight

Provides the HEFTY weight for this task.