Pandas interpolate() backwards in dataframe

It seems it works only with parameter limit see docs [In 47]:

Add a limit_direction keyword argument that works with limit to enable interpolate to fill NaN values forward, backward, or both (GH9218, GH10420, GH11115)

records = pd.DataFrame({'name': {0: 'a', 1: 'a', 2: 'a', 3: 'a', 4: 'a', 5: 'a', 6: 'a', 7: 'a', 8: 'a', 9: 'a'}, 'days': {0: 0.0, 1: np.nan, 2: np.nan, 3: np.nan, 4: 4.0, 5: 5.0, 6: np.nan, 7: np.nan, 8: np.nan, 9: 9.0}}, columns=['name','days'])print (records)  name  days0    a   0.01    a   NaN2    a   NaN3    a   NaN4    a   4.05    a   5.06    a   NaN7    a   NaN8    a   NaN9    a   9.0

#by default limit_direction='forward'records['forw'] = records['days'].interpolate(method='linear',                                               limit=1)records['backw'] = records['days'].interpolate(method='linear',                                               limit_direction='backward',                                                limit=1)records['both'] = records['days'].interpolate(method='linear',                                               limit_direction='both',                                               limit=1)print (records)  name  days  forw  backw  both0    a   0.0   0.0    0.0   0.01    a   NaN   1.0    NaN   1.02    a   NaN   NaN    NaN   NaN3    a   NaN   NaN    3.0   3.04    a   4.0   4.0    4.0   4.05    a   5.0   5.0    5.0   5.06    a   NaN   6.0    NaN   6.07    a   NaN   NaN    NaN   NaN8    a   NaN   NaN    8.0   8.09    a   9.0   9.0    9.0   9.0