_datetime_nanmin() broken (TypeError)

[RondeauG]

There appears to be either a typo (min instead of np.nanmin) or a missing import in the new _datetime_nanmin function.

Specifically, line 363:
array = min(array, skipna=True)

Always results in:
TypeError: 'skipna' is an invalid keyword argument for min()

Permalink:
https://github.com/xarray-contrib/flox/blob/4f6164f74a657d95f795c374e31e91a7e7643e9f/flox/xrutils.py#L363C5-L363C36

[dcherian]

dang it. Can you provide an example please? Clearly the current test suite isn't good enough.

[RondeauG]

This is failing for me with flox v0.9.14 and xarray v2024.9.0.

ds = xr.tutorial.open_dataset("air_temperature")
ds.time.resample(time="D").count()

EDIT: An example using flox directly:

import flox
import numpy as np
array = np.array([np.datetime64("2000-01-01"), np.datetime64("2000-01-02"), np.datetime64("2000-01-03")])
by = [np.array([0, 0, 1])]
flox.groupby_reduce(array, by, func="nanmin")