The two functions share underlying logic with the following differences:
process_I_vector supports incidence2 whereas process_I does notprocess_I_vector does not transfer local[1] to imported if local[1] > 0.
Finally process_I_vector will flatten inputs to a vector.
If not transferring local[1] to imported if local[1] > 0 is a choice, it should be flagged in the documentation.
On the other hand, if it is an oversight we should definitely fix it.
But even if it is the case that not treating local[1] to imported if local[1] > 0 is a choice, we could easily get rid of duplicated code and checks by adding a flag to process_I and having process_I_vector simply flatten the output of process_I
Thoughts @annecori and @rebeccanash ?
The two functions share underlying logic with the following differences:
process_I_vectorsupports incidence2 whereas process_I does notprocess_I_vectordoes not transfer local[1] to imported if local[1] > 0.Finally
process_I_vectorwill flatten inputs to a vector.If not transferring local[1] to imported if local[1] > 0 is a choice, it should be flagged in the documentation.
On the other hand, if it is an oversight we should definitely fix it.
But even if it is the case that not treating local[1] to imported if local[1] > 0 is a choice, we could easily get rid of duplicated code and checks by adding a flag to
process_Iand havingprocess_I_vectorsimply flatten the output ofprocess_IThoughts @annecori and @rebeccanash ?