Closed amitschang closed 4 months ago
All modified and coverable lines are covered by tests :white_check_mark:
Project coverage is 100.00%. Comparing base (
05aa2e0
) to head (3d6ead8
).
:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.
Should the script also include tests to verify that updating or overriding individual keys within a nested dictionary configuration does not inadvertently affect adjacent keys?
@xiangchenjhu, could you provide an example? Do you mean like:
given task function like:
def task(indf, dict_kwarg: dict, ...):
that this sets dict_kwarg['dict_elem']
and leaves other keys alone?
--set-config task.kwargs.dict_kwarg.dict_elem = 1
I don't think at the moment it descends that far. But maybe you mean something else?
Your example captures my concern. It's understandable if supporting such deep configuration updates isn't the goal for this change
@xiangchenjhu, cool thanks. You brought up an interesting prospect though, so I made issue to track that possibility: https://github.com/ssec-jhu/dplutils/issues/79
Previously dict would be updated at the base-level meaning any parameters at level below task would be overridden. Here use the executor set_config method which already handles dict update at appropriate depth.
Resolves https://github.com/ssec-jhu/dplutils/issues/77