I would appreciate any edification on the Mypy failures. Looking at the indicated lines, i'm 🤷 .

I solved the mypy errors in a highly dubious way. 👀

In the off-hand chance this is reviewed before I push again, do not merge. I have a fix to encodings not getting properly roundtripped in Ds.from_dict(ds.to_dict). it was minor to fix but making sure it's tested will take a min

@dcherian thanks! I didnt incoroprate any suggestions yet.
regarding the inequality of encodings of datasets is obscured by assert_identical(a, b) not evaluating encodings. it seems like it should have an option to also check encodings (or not).

i kinda implied, but I'll just state that the extra code to test equality of encodings is not handsome.

I'm happy to "fix" the mypy issues, but it's on that I suspect might be requested for changes (if I recall correctly, it's just in the tests)

Copying my comment from #1599 (comment)

Perhaps we should have array_to_list: bool instead. If False, we just preserve the underlying array type.
Then the user could do ds.as_numpy().to_dict(array_to_list=False) to always get numpy arrays as #7739

array_data or data_as_array could be other options

Making all the requested changes, the above should resolve momentarily.

I like this "trick"/suggestion:

And a design question/suggestion: what about instead of adding another kwarg, you could use data = True / False / "numpy"?

I will implement this if we are in agreement with @dcherian

what about instead of adding another kwarg, you could use data = True / False / "numpy"?

Oh yeah, I like this. Only suggestion is data = True / False / "array" / "list" where True and "list" are synonymous.

I followed data = True / False / "array" / "list"

Thanks @jmccreight great work!