REGR: change in Series.astype(str) behavior for None

[itholic]

Question about pandas

Let's say we have a Series with None like the below.

pser = pd.Series(["hi", "hi ", " ", " \t", "", None], name="x")

print(pser)
0      hi
1     hi
2
3      \t
4
5    None
Name: x, dtype: object

the last value is None, but it is casted to nan after using astype with str parameter.

print(pser.astype(str))
0     hi
1    hi
2
3     \t
4
5    nan
Name: x, dtype: object

Is it intended behavior in pandas 1.1.1. ??

Thanks :)

[dsaxton]

Thanks @itholic, this seems like a regression and is due to #35519

cc @topper-123

[itholic]

Thanks, @dsaxton :)