issues

• [x] follow-up to #7415

It seems in a previous PR I "temporarily" pinned numpy to get CI to pass, but then forgot to unpin later and merged it as-is. As a result, we have not been running the main CI with numpy>=1.24 ever since, even though now numpy=1.25 has been around for a while.

As mentioned in https://github.com/pydata/xarray/issues/7716#issuecomment-1497623839, this tries to unpin pandas.

In #5845, we accidentally broke the detection of the backends. Since this has a big impact we probably need to release v0.20.1 very soon.

I'm not sure if it's possible to add tests for this, though.

• [x] Closes #5930
• [ ] Tests added
• [x] Passes pre-commit run --all-files
• [x] User visible changes (including notable bug fixes) are documented in whats-new.rst

As discussed this extends keep_attrs in apply_ufunc to also accept merge strategy names and use merge_attrs (making it easier to add more merge strategies).

Handling attrs on variables / coords is still missing.

• [x] Tests added
• [x] Passes pre-commit run --all-files
• [x] User visible changes (including notable bug fixes) are documented in whats-new.rst

This prepares our docstrings for the improvements to napoleon in the upcoming sphinx release (~I'm not aware of a fixed date, though~ it's going to be released around Aug 08, so in about a week from now). napoleon now has a stricter syntax for parameter type specs with napoleon_use_param = True, which allows it to link to all referenced types.

We can't do much about the inherited docstrings, though, so we'd probably have to fix that upstream.

• [ ] works towards closing #3370
• [x] Passes isort . && black . && mypy . && flake8
• [ ] User visible changes (including notable bug fixes) are documented in whats-new.rst
• [ ] New functions/methods are listed in api.rst

This PR adds tests that cover the module level functions of the public API, similar to #3238 and #3447. - [x] Tests added - [x] Passes black . && mypy . && flake8 - [x] Fully documented, including whats-new.rst for all changes and api.rst for new API

As a reference for myself, these are the functions listed by the docs: * [x] apply_ufunc * [x] align * [x] broadcast * [x] concat * [x] merge * [x] combine_by_coords * [x] combine_nested * [ ] <strike>auto_combine (deprecated)</strike> * [x] masking / selecting: where * [x] replication: full_like, ones_like, zeros_like * [x] dot * [ ] <strike>map_blocks</strike>

Functions not covered by this PR: * auto_combine (deprecated) * map_blocks (dask specific, should be the same as apply_ufunc without dask)

As a follow-up to #3238, this adds tests for datasets. Replacing assert_equal_with_units with assert_identical and adding tests for the toplevel functions are not included, these will be new PRs. - [x] Tests added - [x] Passes black . && mypy . && flake8

As a reference for myself, this is the list of methods from the documentation of Dataset: - [x] creation: Dataset() - [x] contents: copy, assign, assign_coords, assign_attrs, pipe, merge, rename, rename_vars, rename_dims, swap_dims, expand_dims, drop, drop_dims, set_coords, reset_coords - [x] comparisons: equals, broadcast_equals, identical - [x] indexing: loc, isel, sel, head, tail, thin, squeeze, interp, interp_like, reindex, reindex_like, set_index, reset_index, reorder_levels - [x] missing value handling: isnull, notnull, combine_first, count, dropna, fillna, ffill, bfill, interpolate_na, where, isin - [x] computation: apply, reduce, groupby, groupby_bins, rolling, rolling_exp, coarsen, resample, diff, quantile, differentiate, integrate - [x] aggregation: all, any, argmax, argmin, max, min, mean, median, prod, sum, std, var - [x] ndarray methods: astype, argsort, clip, conj, conjugate, imag, round, real, cumsum, cumprod, rank - [x] grouped operations: assign, assign_coords, first, last, fillna, where, quantile - [x] reshaping and reorganizing: transpose, stack, unstack, to_stacked_array, shift, roll, sortby

these methods are not covered: - creation: decode_cf - dictionary interface: __getitem__, __setitem__, __delitem__, update, items, values - broadcast_like (since I also forgot this for DataArray, I'll create a new PR for it)