[Python] NaT date32 values are only converted to nulls if from_pandas is used

[asfimport]

np.array([None, date(2017, 4, 4)], dtype='datetime64[D]')
pa.array(expected, from_pandas=True) -> [null, 2017-4-4]
pa.array(expected, from_pandas=True) -> [1970-1-1, 2017-4-4]

Reporter: Uwe Korn / @xhochy
Assignee: Wes McKinney / @wesm

PRs and other links:

• GitHub Pull Request #1399

_{Note: This issue was originally created as ARROW-1891. Please see the migration documentation for further details.}

[asfimport]

Uwe Korn / @xhochy:
@wesm is this expected behaviour or a bug? I could not infer that from the comment on pa.array().

[asfimport]

Wes McKinney / @wesm:
Seems buggy. I will take a look!

[asfimport]

Wes McKinney / @wesm:
Issue resolved by pull request 1399
#1399