BUG: Prevent pd.Series.groupby from showing FutureWarning

[sanggon6107]

• closes BUG: pd.Series.groupby issues FutureWarning #60831
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

Linked Issue

• Future warning observed when series.groupby is called where its name is an integer. The warning seems to stem from is_in_obj() defined in get_grouper() ; Below indexing results in the warning since gpr.name is integer type. For example :

>>>import pandas as pd
>>>arr = [['a', 'a', 'b', 'b'], ['aa', 'bb', 'aa', 'bb']]
>>>idx = pd.MultiIndex.from_arrays(arr, names=['index1', 'index2'])
>>>ser = pd.Series([10, 20, 30, 40], index=idx, name=2)
>>>ser

index1  index2
a       aa        10
        bb        20
b       aa        30
        bb        40
Name: 2, dtype: int64

>>>ser.groupby(ser)
FutureWarning: Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`

Proposed Code Change

• Since pd.Series doesn't have columns, added type check so that is_in_obj() is called only when obj is a DataFrame.

    for gpr, level in zip(keys, levels):
        if isinstance(obj, DataFrame) and is_in_obj(gpr):  # df.groupby(df['name'])
            in_axis = True
            exclusions.add(gpr.name)

Please let me know if there are any changes needed. Thanks!

[rhshadrach]

lgtm. Since the behavior isn't really changing here (other than to avoid a warning in 2.3.x that no longer exists on main), I think we're okay without a test.

[rhshadrach]

The failures here don't look related, but I'm not seeing them fail in other PRs. @mroeschke - have you seen test_apply_raw_float_frame_lambda fail in other PRs recently?

[mroeschke]

Thanks @sanggon6107

Hmm interesting, no I haven't seen that test fail recently. I wonder if it's only happening on the 2.3.x branch for some reason