BUG: corner cases in DTI.get_value, Float64Index.get_value #31163
Conversation
| @@ -773,3 +773,14 @@ def test_get_loc_nat(self): | |||
| # GH#20464 | |||
| index = DatetimeIndex(["1/3/2000", "NaT"]) | |||
| assert index.get_loc(pd.NaT) == 1 | |||
|
|
|||
There was a problem hiding this comment.
maybe better to parmateize over the accepted nulls (do we have a fixture for date time nulls?) and then have an errors version
There was a problem hiding this comment.
ill put that on my list of follow-ups, we're about 2/3 of the way towards having get_value be internally consistent.
Thoughts on the Series.at outlier?
There was a problem hiding this comment.
what’s the Series.at outlier?
There was a problem hiding this comment.
ill put that on my list of follow-ups, we're about 2/3 of the way towards having get_value be internally consistent.
sure
There was a problem hiding this comment.
what’s the Series.at outlier?
in the test here ser.loc[4] behaves like ser.loc[4.0], while ser.at[4] raises
jreback
added
the
Indexing
| @@ -773,3 +773,14 @@ def test_get_loc_nat(self): | |||
| # GH#20464 | |||
| index = DatetimeIndex(["1/3/2000", "NaT"]) | |||
| assert index.get_loc(pd.NaT) == 1 | |||
|
|
|||
There was a problem hiding this comment.
ill put that on my list of follow-ups, we're about 2/3 of the way towards having get_value be internally consistent.
sure
|
gentle ping; this a blocker before fixing the .at bug |
Series lookups are affected for the Float64Index case.