⚠️ Nightly upstream-dev CI failed ⚠️

[github-actions]

Workflow Run URL

Python 3.9 Test Summary Info

FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-False-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-sum-True-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-False-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-None-prod-True-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-False-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-sum-True-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-False-bool_-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-float-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-float-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-int-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-int-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-float32-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-float32-2]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-bool_-1]
FAILED xarray/tests/test_duck_array_ops.py::test_min_count[None-True-x-prod-True-bool_-2]

[mathause]

I think this is an upstream issue (or intended change).

import numpy as np
import pandas as pd

pd.Series([1.7, 0.5, np.nan]).sum(skipna=None)

returns 2.2 in pandas 1.3.4 and nan in main. Pandas is used to verify the result. That means that result is still correct in the test but expected has changed.

xarray/xarray/tests/test_duck_array_ops.py

Line 580 in fdabf3b

expected = series_reduce(da, func, skipna=skipna, dim=aggdim, min_count=min_count)

uses pandas

xarray/xarray/tests/test_duck_array_ops.py

Lines 262 to 263 in fdabf3b

	se = da.to_series()
	return from_series_or_scalar(getattr(se, func)(**kwargs))

I'll go check if this is reported upstream.

[mathause]

See pandas-dev/pandas#44178