Skip to content

re-order add_feature arguments #633

@DaniBodor

Description

@DaniBodor

The function is currently defined by default as follows:

def add_features(
    pdb_path: str,  # noqa: ARG001
    graph: Graph,
    single_amino_acid_variant: SingleResidueVariant | None = None,  # noqa: ARG001
) -> None:

There is only 1 feature for which the pdb_path is actually used. However as currently implemented it must always be given. For this reason, it would make more sense if pdb_path is requested last, so that it can just be omitted in most add_features calls.

But I think it would be even better to always take **kwargs as an argument in this function. Any specific argument needed for the feature in question can be explicitly named, and all others can be "hidden" in the kwargs. This also opens up the possibility for adding other keywords to future (or user defined) features that are not currently used, while maintaining a default structure.

Metadata

Metadata

Assignees

No one assigned

    Labels

    covariatesmodules from the 'features' subpackagediscDiscussion needed, to better define the task/sstaleissue not touched from too much time

    Type

    No type

    Projects

    Status

    To do

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions