Move parallelcompat and chunkmanagers to NamedArray

[TomNicholas]

@dcherian I got to this point before realizing that simply moving parallelcompat.py over isn't what it says in the design doc, which instead talks about

• Could this functionality be left in Xarray proper for now? Alternative array types like JAX also have some notion of "chunks" for parallel arrays, but the details differ in a number of ways from the Dask/Cubed.
• Perhaps variable.chunk/load methods should become functions defined in xarray that convert Variable objects. This is easy so long as xarray can reach in and replace .data

I personally think that simply moving parallelcompat makes sense so long as you expect people to use chunked NamedArray objects. I see the chunked arrays as special cases of duck arrays, and my understanding is that NamedArray is supposed to have full support for duckarrays.

cc @andersy005

• As requested in NamedArray tracking issue #8238
• Tests added
• User visible changes (including notable bug fixes) are documented in whats-new.rst
• New functions/methods are listed in api.rst

[andersy005]

thank you for doing this work, @TomNicholas! moving these modules unblock me.

I personally think that simply moving parallelcompat makes sense so long as you expect people to use chunked NamedArray objects. I see the chunked arrays as special cases of duck arrays, and my understanding is that NamedArray is supposed to have full support for duckarrays.

👍🏽 i concur. and yes, NamedArray will have full support for duckarrays.

[TomNicholas]

Moving utils.either_dict_or_kwargs to namedarray.utils required touching a lot of files to fix!

Also moving the entrypoint seems to have broken the discovery of the DaskManager at least in my local environment, let's see what happens in the CI...

[TomNicholas]

Also moving the entrypoint seems to have broken the discovery of the DaskManager at least in my local environment, let's see what happens in the CI...

This works in the CI! But it still breaks my local environment 🙁 I suspect this means it will break other developer's local environments too. The problem is these zombie entrypoints that I cannot seem to get rid of (even after deleting the __pycache__ directory). There should just be one in this list:

In [1]: from importlib.metadata import entry_points

In [2]: entry_points().get("xarray.chunkmanagers", ())
Out[2]: 
(EntryPoint(name='dask', value='xarray.namedarray.daskmanager:DaskManager', group='xarray.chunkmanagers'),
 EntryPoint(name='dask', value='xarray.core.daskmanager:DaskManager', group='xarray.chunkmanagers'),
 EntryPoint(name='dask', value='xarray.namedarray.daskmanager:DaskManager', group='xarray.chunkmanagers'))

[andersy005]

this is ready for another round of review :)

[TomNicholas]

I can't approve my own PR, but I approve! 👍