New defaults for `concat`, `merge`, `combine_*` #10062

jsignell · 2025-02-19T22:17:11Z

Replaces #10051

Towards Stricter defaults for concat, combine, open_mfdataset, merge #8778
- after a sufficiently long deprecation cycle there should be another PR that:
  - removes the option (but doesn't throw if people are still setting it)
  - removes all the FutureWarnings
- after even more time one last PR that:
  - encodes the new defaults right into the function signature
  - removes the extra warnings that throw on failure
xref Merging coordinates computes array values #9481
xref slow performance with open_mfdataset #1385
Tests added
User visible changes (including notable bug fixes) are documented in whats-new.rst
~~New functions/methods are listed in api.rst~~

This PR attempts to throw warnings if and only if the output would change with the new kwarg defaults. To exercise the new default I am toggling the option back and forth and running the existing test suite.

With new kwarg values

Run all the tests with use_new_combine_kwarg_defaults=True -> 78 failed
Change to use_new_combine_kwarg_defaults=False and run the last failed -> 76 failed, 2 passed

Those 2 are missed alarms. In these two cases the behavior will be different when xarray switched to using new default kwarg values and there is no warning. But I am not totally sure whether we need to handle them because they are tests for conditions that used to raise an error and with the new defaults they do not.

FAILED xarray/tests/test_concat.py::TestConcatDataset::test_concat_do_not_promote - Failed: DID NOT RAISE <class 'ValueError'>
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_error - Failed: DID NOT RAISE <class 'xarray.core.merge.MergeError'>

With old kwarg values

Run all the tests with use_new_combine_kwarg_defaults=False -> 119 failed
Change to use_new_combine_kwarg_defaults=True and run the last failed -> 76 failed, 43 passed

Those 44 are potentially false alarms - they could also be tests that just don't assert that the result exactly matches some fixed expectation - but mostly they are false alarms.

Here is a list of them

FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[outer-minimal-nested-t] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[outer-minimal-by_coords-None] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[inner-minimal-nested-t] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[inner-minimal-by_coords-None] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[left-minimal-nested-t] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[left-minimal-by_coords-None] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[right-minimal-nested-t] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_does_same_as_concat[right-minimal-by_coords-None] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataset_combine_attrs[drop] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataset_combine_attrs[override] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataset_combine_attrs[no_conflicts] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataset_combine_attrs[identical] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataset_combine_attrs[drop_conflicts] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataset_attr_by_coords - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_open_mfdataset_dataarray_attr_by_coords - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_common_coord_when_datavars_minimal - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestOpenMFDatasetWithDataVarsAndCoordsKw::test_invalid_data_vars_value_should_fail - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_backends.py::TestDask::test_open_mfdataset_concat_dim_none - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::TestDask::test_open_mfdataset_concat_dim_default_none - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_backends.py::test_h5netcdf_storage_options - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_combine.py::TestNestedCombine::test_auto_combine_2d - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_combine.py::TestNestedCombine::test_auto_combine_2d_combine_attrs_kwarg - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_combine.py::TestCombineDatasetsbyCoords::test_infer_order_from_coords - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_combine.py::TestCombineDatasetsbyCoords::test_combine_by_coords_no_concat - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_concat.py::TestConcatDataset::test_concat_coords_kwarg[dim1-minimal] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_concat.py::TestConcatDataset::test_concat_coords_kwarg[dim1-all] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_concat.py::TestConcatDataset::test_concat_errors - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_concat.py::TestConcatDataArray::test_concat_avoids_index_auto_creation - FutureWarning: In a future version of xarray the default value for coords will change from coords='different' to coords='minimal'. This is likely to lead to different results when multiple datasets have m...
FAILED xarray/tests/test_dask.py::test_map_blocks_da_ds_with_template[obj1] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_dask.py::test_map_blocks_errors_bad_template[obj1] - FutureWarning: In a future version of xarray the default value for data_vars will change from data_vars='all' to data_vars='minimal'. This is likely to lead to different results when multiple datasets hav...
FAILED xarray/tests/test_dataset.py::TestDataset::test_dataset_math_auto_align - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_datasets - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[no_conflicts-attrs10-attrs20-expected_attrs0-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[no_conflicts-attrs11-attrs21-expected_attrs1-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[no_conflicts-attrs12-attrs22-expected_attrs2-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[no_conflicts-attrs13-attrs23-expected_attrs3-True] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[drop-attrs14-attrs24-expected_attrs4-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[identical-attrs15-attrs25-expected_attrs5-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[identical-attrs16-attrs26-expected_attrs6-True] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[override-attrs17-attrs27-expected_attrs7-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[drop_conflicts-attrs18-attrs28-expected_attrs8-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_arrays_attrs_variables[<lambda>-attrs19-attrs29-expected_attrs9-False] - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...
FAILED xarray/tests/test_merge.py::TestMergeFunction::test_merge_no_conflicts_preserve_attrs - FutureWarning: In a future version of xarray the default value for compat will change from compat='no_conflicts' to compat='override'. This is likely to lead to different results when combining overlappin...

About half of them are triggered by my attempt to catch cases where different datasets have matching overlapping variables and with compat='no_conflicts' you might get different results than with compat='override'. There might be a better way, but we want to be careful to avoid calling compute.

Updating tests

After investigating the impact on existing tests, I updated the tests to make sure they still test what they were meant to test by passing in any required kwargs and I added new tests that explicitly ensure that for a bunch of different inputs, using old defaults throws a warning OR output is the same with new and old defaults.

Notes

I used sentinel values to indicate a kwarg value that the user has not explicitly set. The benefit of using this approach is that it makes it more obvious that people should not be setting these kwargs to None in their own code. Also with this approach it is possible to encode both the old and the new default value in the function signature.
When users have opted in to the new default values there is logic that tries to catch any errors that might be related to the new default kwargs and adds that context to the error message. In case people don't pay attention to the warnings this will at least give them some indication that things have changed and explain new errors that crop up. This only applies when there is already an error, so they don't catch places where the code succeeds but the result is different than before.

jsignell · 2025-02-19T22:23:06Z

xarray/core/groupby.py

+                coords="different",
+                compat="equals",
+                join="outer",
+            )


I hard-coded these to the old defaults since there is no way for the user to set them.

I agree with this approach. These options result in confusing groupby behaviour (#2145) but we can tackle that later

jsignell · 2025-02-26T19:21:30Z

The last test file that I need to work on is test_concat.py

dcherian

Thanks for taking this on!

I'd like to avoid adding the extra option. Can you remind us of what that would be useful please?

dcherian · 2025-03-03T19:08:05Z

xarray/core/groupby.py

+                coords="different",
+                compat="equals",
+                join="outer",
+            )


I agree with this approach. These options result in confusing groupby behaviour (#2145) but we can tackle that later

xarray/core/dataset.py

xarray/core/concat.py

xarray/core/parallel.py

xarray/tests/test_merge.py

dcherian · 2025-03-03T20:08:45Z

xarray/tests/test_merge.py

        assert_identical(ds2, actual)

-        actual = ds2.merge(ds1)
+        actual = ds2.merge(ds1, compat="override")


ok this is dangerous, the dimensionality of 'x' in the output depends on the order of merging.

I intended compat="override" to be used when "x" is the same in all datasets, not merely 'compatible' as is here. It seems like we would want to at minimum, assert that ndim is same for all x in all datasets.

Opened #10094

xarray/tests/test_merge.py

xarray/core/concat.py

jsignell · 2025-03-04T15:41:42Z

I'd like to avoid adding the extra option. Can you remind us of what that would be useful please?

The option is part of the deprecation process. Basically how it works is in this PR the default values for these kwargs do not change. BUT you get FutureWarnings if the new default values would lead to a different result or an error. The use_new_combine_kwarg_defaults (this absolutely does not have to be the final name) allows you to opt into the new defaults prematurely. Setting that option to True lets you encounter any issues that might arise and it turns off the FutureWarnings.

This is how I'm thinking about the plan after this PR lands:

after a sufficiently long deprecation cycle there should be another PR that:
- removes the option (but (probably) doesn't throw if people are still setting it)
- removes all the FutureWarnings
after even more time one last PR that:
- encodes the new defaults right into the function signature
- removes the extra context about new defaults that is included in some error messages

I guess you could get rid of the option the tradeoff is that to get rid of the FutureWarnings people may have to hardcode to the new defaults. But there aren't too many false alarms (~~I need to update the PR description~~ UPDATED) so most of the time when there is a FutureWarning it does indicate that hardcoding the kwarg value is a good idea.

…atarrays

jsignell · 2025-03-04T21:02:00Z

This is ready for review! The failing test looks like it is also failing on main.

jsignell · 2025-03-07T21:19:18Z

Wellll maybe the unit tests will pass now. I'll fix mypy and the doctests next week.

jsignell · 2025-03-10T18:02:05Z

The mypy failures now match those in other PRs - I opened #10110 to track

xarray/tests/test_combine.py

dcherian · 2025-07-19T04:28:45Z

xarray/structure/combine.py

+        except AlignmentError as e:
+            e.add_note(
+                "If you are intending to concatenate datasets, please specify the concatenation dimension explicitly. "
+                "Using merge to concatenate is quite inefficient."


So happy we can disallow this nonsense!

I didn't know about e.add_note! That's very nice.

for more information, see https://pre-commit.ci

jsignell · 2025-07-22T15:25:44Z

Thanks Julia! I powered through test_concat and test_merge. I'll need another 2 hours to do test_backends and test_combine. But hopefully we can get this in for the next release.

It all looks good so far, I managed to reduce the false positive warnings on merge, so that's a good win!

That is awesome! Thanks so much for taking a look. It's great to have fewer false positives! Let me know if there is anything I can do to help!

dcherian · 2025-07-27T02:18:01Z

xarray/structure/concat.py

@@ -100,14 +108,18 @@ def concat(
            load the data payload of data variables into memory if they are not
            already loaded.
          * "all": All data variables will be concatenated.
+          * None: Means ``"all"`` if ``dim`` is not present in any of the ``objs``,


This is new.

We want to distinguish between concatenating ålong an existing dimension (numpy's concatenate) and concatenating along a new dimension (numpy's stack).

Without this (and with data_vars="minimal", coords="minimal" as default), we are not doing any concatenation if asked to concatenate along a new dim. That behaviour is certainly wrong. So now the new default we are migrating to is data_vars=None, coords="minimal" where we will at least stack the data vars when ask to concat along a new dimension

Ah ok so effectively make the kwarg conditional on the presence of dim in the dataset? I guess that's ok because it is discernible without loading data?

yes the improvement is that we don't ever check the data to make the decision. And because of named dimensions, we can distinguish between concatenate and stack (in numpy terms).

I'll bring this up for discussion at Wednesday's meeting and then merge if there are no objections.

Thanks for your work here!

jsignell · 2025-07-28T14:07:53Z

doc/whats-new.rst

  - ``coords``: "minimal"
  - ``compat``: "override"
  - ``join``: "exact"

-  By `Julia Signell <https://github.com/jsignell>`_.
+  (:issue:`8778`, :issue:`1385`, :pull:`10062`). By `Julia Signell <https://github.com/jsignell>`_.


You should probably add yourself as an author on this one 😅 @dcherian

jsignell

There might just be a few cleanups left with data_vars=None as the default. I can push if that is helpful, just don't want to step on your toes if you are still going :)

xarray/core/options.py

jsignell · 2025-07-28T14:12:02Z

xarray/structure/combine.py

+        except AlignmentError as e:
+            e.add_note(
+                "If you are intending to concatenate datasets, please specify the concatenation dimension explicitly. "
+                "Using merge to concatenate is quite inefficient."


I didn't know about e.add_note! That's very nice.

xarray/structure/concat.py

dcherian · 2025-07-28T16:20:27Z

I can push if that is helpful,

Absolutely, please go ahead.

I wasn't sure that you were still interested in pushing this over line given how long this PR has been open :) . I can review again when you're done

jsignell · 2025-07-28T16:45:14Z

Hahah yeah still interested! I just didn't know what to do to make it palatable :) so I'm stoked to have your review!

jsignell · 2025-07-28T18:22:08Z

Ok! I pushed the changes I was talking about and merged in main. I think the flaky tests are just being flaky 🤷🏻

dcherian · 2025-07-28T19:01:00Z

I just didn't know what to do to make it palatable :)

Hehe, it's just a hard complex change. I had to check it out and use a debugger on a few tests to fully grok things, which then led to the fixes I pushed.

jsignell · 2025-07-28T19:43:29Z

I just didn't know what to do to make it palatable :)

Hehe, it's just a hard complex change. I had to check it out and use a debugger on a few tests to fully grok things, which then led to the fixes I pushed.

Oh totally! People should take their time to feel comfortable with it.

Remove default values in private functions

5c56acf

jsignell mentioned this pull request Feb 19, 2025

Set up new default values for kwargs on open_mfdataset #10051

Closed

4 tasks

jsignell commented Feb 19, 2025

View reviewed changes

Use sentinel value to change default with warnings

5461a9f

jsignell force-pushed the concat_default_kwargs branch from 0e65034 to 5461a9f Compare February 24, 2025 20:07

jsignell added 5 commits February 24, 2025 15:51

Remove unnecessary warnings

e16834f

Use old kwarg values within map_blocks, concat dataarray

9c50125

Merge branch 'main' into concat_default_kwargs

b0cf17a

Switch options back to old defaults

0026ee8

Update tests and add new ones to exercise options

4d4deda

dcherian reviewed Mar 3, 2025

View reviewed changes

jsignell added 2 commits March 4, 2025 09:15

Merge branch 'main' into concat_default_kwargs

5a4036b

Use emit_user_level_warning rather than warnings.warn

912638b

jsignell commented Mar 4, 2025

View reviewed changes

xarray/core/concat.py Outdated Show resolved Hide resolved

jsignell mentioned this pull request Mar 4, 2025

Tracking internal uses of concat that don't allow setting all the kwargs #10095

Open

jsignell added 4 commits March 4, 2025 11:11

Change hardcoded defaults

67fd4ff

Fix up test_concat

4f38292

Add comment about why we allow data_vars='minimial' for concat over d…

51ccc89

…atarrays

Tidy up tests based on review

aa3180e

jsignell marked this pull request as ready for review March 4, 2025 21:01

jsignell mentioned this pull request Mar 5, 2025

Stricter defaults for concat, combine, open_mfdataset, merge #8778

Open

Merge branch 'main' into concat_default_kwargs

93d2abc

jsignell added 3 commits March 10, 2025 12:58

Trying to resolve mypy issues

e517dcc

Fix mypy in tests

0e678e5

Fix doctests

37f0147

dcherian reviewed Jul 18, 2025

View reviewed changes

xarray/tests/test_combine.py Show resolved Hide resolved

Clean up combine

375895a

dcherian force-pushed the concat_default_kwargs branch from b683b37 to 6e0e0eb Compare July 19, 2025 04:26

dcherian reviewed Jul 19, 2025

View reviewed changes

dcherian added 2 commits July 18, 2025 22:35

Cleanup test_backends

1d7da8d

Fix mypy

f63ed66

dcherian force-pushed the concat_default_kwargs branch from 6e0e0eb to f63ed66 Compare July 19, 2025 04:36

pre-commit-ci bot and others added 3 commits July 19, 2025 04:36

[pre-commit.ci] auto fixes from pre-commit.com hooks

af8b8a2

for more information, see https://pre-commit.ci

one more

248d53a

last one?!

8933245

Adjust meaning of "minimal" when concatenating along a new dim

bec25f6

dcherian force-pushed the concat_default_kwargs branch from 69b9ef6 to bec25f6 Compare July 26, 2025 23:53

dcherian reviewed Jul 27, 2025

View reviewed changes

Add data_vars=None as default.

114845f

dcherian force-pushed the concat_default_kwargs branch 2 times, most recently from 29e0da6 to a45ff5f Compare July 27, 2025 02:38

Add types

f0eab2e

dcherian force-pushed the concat_default_kwargs branch from a45ff5f to f0eab2e Compare July 27, 2025 04:21

Update whats-new

292fe87

jsignell commented Jul 28, 2025

View reviewed changes

jsignell added 3 commits July 28, 2025 13:21

Use None as new default for data_vars

e4e00ba

Merge branch 'main' into concat_default_kwargs

69879c4

Convince mypy that None is a valid data_vars value

9b435c4

Uh oh!

New defaults for concat, merge, combine_* #10062

Are you sure you want to change the base?

New defaults for concat, merge, combine_* #10062

Uh oh!

Conversation

jsignell commented Feb 19, 2025 • edited by dcherian Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

With new kwarg values

With old kwarg values

Updating tests

Notes

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jsignell commented Feb 26, 2025

Uh oh!

dcherian left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

jsignell commented Mar 4, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jsignell commented Mar 4, 2025

Uh oh!

jsignell commented Mar 7, 2025

Uh oh!

jsignell commented Mar 10, 2025

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jsignell commented Jul 22, 2025

Uh oh!

dcherian Jul 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

jsignell left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

dcherian commented Jul 28, 2025

Uh oh!

jsignell commented Jul 28, 2025

Uh oh!

jsignell commented Jul 28, 2025

Uh oh!

dcherian commented Jul 28, 2025

Uh oh!

jsignell commented Jul 28, 2025

Uh oh!

Uh oh!

New defaults for `concat`, `merge`, `combine_*` #10062

New defaults for `concat`, `merge`, `combine_*` #10062

jsignell commented Feb 19, 2025 •

edited by dcherian

Loading

jsignell commented Mar 4, 2025 •

edited

Loading

dcherian Jul 27, 2025 •

edited

Loading