Refactor use of the product datamode in GWCSDrizzle

[stscijgbot-jp]

Issue JP-2973 was created on JIRA by Jonathan Eisenhamer:

The API of GWCSDrizzle could use with a bit of a refactor. The product DataModel could be better integrated, started in PR#⁠7295, handling the rest of the data arrays in similar fashion.

There is also redundant initialization to None for many of the internal arrays.

Finally, the final output of GWCSDrizzle is unclear. It is in the input product itself. This may be just a documentation issue, though one may consider a method/property that explicitly queries for the result.

[stscijgbot-jp]

Comment by David Law on JIRA:

Jonathan Eisenhamer More long term technical debt?

[stscijgbot-jp]

Comment by Jonathan Eisenhamer on JIRA:

yep, including any issues with this question. may actually consider putting a "debt" label on issues like this. ultimately up to project leads for this type of work. busy work for interns if needed?

[stscijgbot-jp]

Comment by David Law on JIRA:

I was thinking some kind of more accurate label might be helpful.  I've been using 'Trivial' for technical debt, but not only is that not an obvious use, it's at times patently untrue.  Will discuss.

[stscijgbot-jp]

Comment by Nadia Dencheva on JIRA:

The interface to the drizzle C code was refactored and this will be reflected in jwst.resample.GWCSDrizzle.

#8866

[stscijgbot-jp]

Comment by Nadia Dencheva on JIRA:

The drizzle I/O and "Drizzle" class API was refactored and streamlined:

spacetelescope/drizzle#134

The corresponding PR in the JWST pipeline is:

#8866