BUG: value counts with mixed NaNs is oddly rendering

[jreback]

in the 1.40rc1 release notes: https://pandas.pydata.org/pandas-docs/version/1.4/whatsnew/v1.4.0.html#null-values-are-no-longer-coerced-to-nan-value-in-value-counts-and-mode

We are seeing distinct NaN , but I think we should be seeing NaN and NaT

Looks like we patched here: #42743

[jreback]

cc @realead if you can

[realead]

@jreback Sorry I didn't pay the needed attention to the output and overlooked that when a series is printed all null-values are printed as NaN, i.e.:

import pandas as pd
import numpy as np
s=pd.Series([1, 2, 3, 4], [True, None, np.nan, pd.NaT])

and now

print(s.keys())
print(s[np.nan])

work as expected:

Index([True, None, nan, NaT], dtype='object')
3

but

print(s)

yields:

True    1
NaN     2
NaN     3
NaN     5

i.e. all NaNs.

For the documentation it is probably better to create another example.

But is this the behavior of series one would like to have/expect with print?

[realead]

I think this issue is closed via #45283