Skip to content

Get local config

Submodule to handle the local-backend configuration for a WorkflowTask

FUNCTION DESCRIPTION
get_local_backend_config

Prepare a specific LocalBackendConfig configuration.

Classes

Functions:

get_local_backend_config(shared_config, wftask, which_type, tot_tasks=1)

Prepare a specific LocalBackendConfig configuration.

The base configuration is the runner-level shared_config object, based on resource.jobs_runner_config. We then incorporate attributes from wftask.meta_{non_parallel,parallel} - with higher priority.

PARAMETER DESCRIPTION

shared_config

Configuration object based on resource.jobs_runner_config.

TYPE: JobRunnerConfigLocal

wftask

WorkflowTaskV2 for which the backend configuration should be prepared.

TYPE: WorkflowTaskV2

which_type

Whether we should look at the non-parallel or parallel part of wftask.

TYPE: Literal['non_parallel', 'parallel']

tot_tasks

Not used here, only present as a common interface.

TYPE: int DEFAULT: 1

RETURNS DESCRIPTION
JobRunnerConfigLocal

A ready-to-use local-backend configuration object.
Source code in fractal_server/runner/executors/local/get_local_config.py 
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
def get_local_backend_config(
    shared_config: JobRunnerConfigLocal,
    wftask: WorkflowTaskV2,
    which_type: Literal["non_parallel", "parallel"],
    tot_tasks: int = 1,
) -> JobRunnerConfigLocal:
    """
    Prepare a specific `LocalBackendConfig` configuration.

    The base configuration is the runner-level `shared_config` object, based
    on `resource.jobs_runner_config`. We then incorporate attributes from
    `wftask.meta_{non_parallel,parallel}` - with higher priority.

    Args:
        shared_config:
            Configuration object based on `resource.jobs_runner_config`.
        wftask:
            WorkflowTaskV2 for which the backend configuration should
            be prepared.
        which_type:
            Whether we should look at the non-parallel or parallel part
            of `wftask`.
        tot_tasks: Not used here, only present as a common interface.

    Returns:
        A ready-to-use local-backend configuration object.
    """

    if which_type == "non_parallel":
        wftask_meta = wftask.meta_non_parallel
    elif which_type == "parallel":
        wftask_meta = wftask.meta_parallel
    else:
        raise ValueError(
            f"Invalid {which_type=} in `get_local_backend_config`."
        )

    __KEY__ = "parallel_tasks_per_job"
    output = shared_config.model_copy(deep=True)
    if wftask_meta and __KEY__ in wftask_meta:
        output.parallel_tasks_per_job = wftask_meta[__KEY__]
    return output