sort seems to sort inplace by default unlike documentation

[s-celles]

Hello,

I noticed

df_ohlc['high'].sort(ascending=False)

raises

ValueError: This Series is a view of some other array, to sort in-place you must create a copy

which seems to mean that, by default, sort uses by default inplace=True

but according to

http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.sort.html

inplace : boolean, default False

Here is a minimal example:

import pandas as pd
import numpy as np
N = 1000
rng = pd.date_range('1/1/2011', periods=N, freq='H')
s = pd.Series(np.random.random(N), index=rng)
df_ohlc = s.resample('1D', how='ohlc')
df_ohlc['high'].sort(ascending=False)

raises ValueError: This Series is a view of some other array, to sort in-place you must create a copy

df_ohlc['high'].sort(ascending=False, inplace=True)

raises same exception

but

df_ohlc['high'].sort(ascending=False, inplace=False)

doesn't raises exception

Kind regards

[jreback]

you are sorting a Series, not a frame
.sort defaults to True for a Series

[s-celles]

Thanks
http://pandas.pydata.org/pandas-docs/stable/generated/pandas.Series.sort.html

it's odd to have by default inplace=True for Series and inplace=False for DataFrame

Now I will remember.

Sorry

[jorisvandenbossche]

It's indeed odd, and there are already a lot of discussions to have this solved (see eg #8239 and #9816), but we can't just change it for backward compatible reasons.