What is your issue?

I am going through the contributor's guide for xarray, and I have a few suggested updates to make the instructions clearer to relatively new contributors like me!

General questions

• If making updates to docstrings, I am unclear if I should use the virtual env xarray-tests or xarray-docs. I assumed I should use xarray-docs since I am only updating docstrings which are fed into the documentation. But this isn't entirely clear, since the file I updated is not in the \docs folder, but is at \xarray\backends\api.rst.
• If only updating docs or docstrings, should I still run pytest locally before pushing? Or do those tests only apply to code updates? Either way, this should be made clear in the contributing guide.

Suggestion updates

• Under Code Formatting:
• Contributors are recommended to use pre-commit via pre-commit install, but when I tried this I found that pre-commit was not installed in the virtual env xarray-docs. It does appear to be installed in virtual env xarray-tests (yml file). Should I run pre-commit when updating docs? If so, we should add pre-commit to virtual env xarray-docs (yml file).
• Under Building the Documentation:
• Add a sentence to make it clear that users can preview the html files from their updated documentation in their local browsers to verify it looks as expected. (This is a minor suggestion, but we might as well be as explicit as possible, especially since docs might more likely be updated by newer contributors.)
• Suggested wording update: "Then you can find the HTML output in the folder xarray/doc/_build/html/. You can preview the html files in your local browser to verify that you see expected behavior based on your changes."

This adds a sentence (below) under "chunks" in doctoring of open_dataset() to clarify current usage for those who want the same behavior as open_zarr() used to provide:

In order to reproduce the default behavior of xr.open_zarr(...) use `xr.open_dataset(..., engine='zarr', chunks={})`.

• [x] Closes #7293

What happened: When running polyfit() on a 3d chunked xarray DataArray, the output is chunked differently than the input array.

What you expected to happen: I expect the output to have the same chunking as the input.

Minimal Complete Verifiable Example: (from @rabernat in https://github.com/xgcm/xrft/issues/116)

Example: number of chunks decreases ```python import dask.array as dsa import xarray as xr

nz, ny, nx = (10, 20, 30) data = dsa.ones((nz, ny, nx), chunks=(1, 5, nx)) da = xr.DataArray(data, dims=['z', 'y', 'x']) da.chunks

-> ((1, 1, 1, 1, 1, 1, 1, 1, 1, 1), (5, 5, 5, 5), (30,))

pf = da.polyfit('x', 1) pf.polyfit_coefficients.chunks

-> ((1, 1, 1, 1, 1, 1, 1, 1, 1, 1), (20,), (30,))

chunks on the y dimension have been consolidated!

pv = xr.polyval(da.x, pf.polyfit_coefficients).transpose('z', 'y', 'x') pv.chunks

-> ((1, 1, 1, 1, 1, 1, 1, 1, 1, 1), (20,), (30,))

and this propagates to polyval

align back against the original data

(da - pv).chunks

-> ((1, 1, 1, 1, 1, 1, 1, 1, 1, 1), (5, 5, 5, 5), (30,))

hides the fact that we have chunk consolidation happening upstream

```

Example: number of chunks increases ```python nz, ny, nx = (6, 10, 4) data = dsa.ones((nz, ny, nx), chunks=(2, 10, 2)) da = xr.DataArray(data, dims=['z', 'y', 'x']) da.chunks

-> ((2, 2, 2), (10,), (2, 2))

pf = da.polyfit('y', 1) pf.polyfit_coefficients.chunks

-> ((2,), (1, 1, 1, 1, 1, 1), (4,))

pv = xr.polyval(da.y, pf.polyfit_coefficients).transpose('z', 'y', 'x') pv.chunks

-> ((1, 1, 1, 1, 1, 1), (10,), (4,))

(da - pv).chunks

-> ((1, 1, 1, 1, 1, 1), (10,), (2, 2))

``` (This discussion started in https://github.com/xgcm/xrft/issues/116 with @rabernat and @navidcy.)

Environment:

Running on Pangeo Cloud