DataFrame.where does not work as expected with inplace=True #7656
Description
According to the documentation, inplace should replace values appropriately with other, but this does not seem to work as expected. This worked in 0.12.
In [6]: df = pandas.DataFrame([{'A': 1, 'B': np.nan, 'C': 'Test'}, {'A': np.nan, 'B': 'Test', 'C': np.nan}]) In [7]: df Out[7]: A B C 0 1 NaN Test 1 NaN Test NaN In [8]: df1 = df.where(~pandas.isnull(df), None) In [9]: df1 Out[9]: A B C 0 1 None Test 1 None Test None In [10]: df.where(~pandas.isnull(df), None, inplace=True) In [11]: df Out[11]: A B C 0 1 NaN Test 1 NaN Test NaN