BUG: 1.1.0 breaks custom indexer support in groupby().rolling()

[WhistleWhileYouWork]

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• (optional) I have confirmed this bug exists on the master branch of pandas.

---

Code Sample, a copy-pastable example

class SimpleIndexer(pd.api.indexers.BaseIndexer):
    ''' Custom `Indexer` duplicating basic fixed length windowing '''
    def get_window_bounds(self, num_values=0, min_periods=None, center=None, closed=None):
        min_periods = self.window_size if min_periods is None else 0
        end = np.arange(num_values, dtype=np.int64) + 1
        start = end.copy() - self.window_size
        #---- Clip to `min_periods`
        start[start < 0] = min_periods
        return (start, end)

x = pd.DataFrame({'a': [1.0,2.0,3.0,4.0,5.0] * 3}, index=[0]*5+[1]*5+[2]*5)
x

Out: 
     a
0  1.0
0  2.0
0  3.0
0  4.0
0  5.0
1  1.0
1  2.0
1  3.0
1  4.0
1  5.0
2  1.0
2  2.0
2  3.0
2  4.0
2  5.0

x.groupby(x.index).rolling(SimpleIndexer(window_size=3), min_periods=1).sum()

Out: 
       a
0 0  1.0
  0  2.0
  0  3.0
  0  4.0
  0  5.0
1 1  1.0
  1  2.0
  1  3.0
  1  4.0
  1  5.0
2 2  1.0
  2  2.0
  2  3.0
  2  4.0
  2  5.0

Problem description

groupby().rolling() does not use custom indexer supplied in window= likely as a result of #34052 where a GroupbyRollingIndexer is used instead.

The output data is always the same as the input regardless of the windowing function.

1.0.5 does not have this behavior.

Expected Output

x.groupby(x.index).rolling(window=3, min_periods=1).sum()

Out: 
        a
0 0   1.0
  0   3.0
  0   6.0
  0   9.0
  0  12.0
1 1   1.0
  1   3.0
  1   6.0
  1   9.0
  1  12.0
2 2   1.0
  2   3.0
  2   6.0
  2   9.0
  2  12.0

Output of pd.show_versions()

INSTALLED VERSIONS ------------------ commit : d9fff27 python : 3.8.2.final.0 python-bits : 64 OS : Linux OS-release : 5.3.0-59-generic Version : #53~18.04.1-Ubuntu SMP Thu Jun 4 14:58:26 UTC 2020 machine : x86_64 processor : x86_64 byteorder : little LC_ALL : None LANG : en_US.UTF-8 LOCALE : en_US.UTF-8

pandas : 1.1.0
numpy : 1.19.1
pytz : 2020.1
dateutil : 2.8.1
pip : 20.2
setuptools : 49.2.1.post20200802
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : 7.14.0
pandas_datareader: None
bs4 : None
bottleneck : None
fsspec : None
fastparquet : None
gcsfs : None
matplotlib : 3.2.1
numexpr : 2.7.1
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : 0.17.0
pytables : None
pyxlsb : None
s3fs : None
scipy : 1.4.1
sqlalchemy : None
tables : 3.6.1
tabulate : None
xarray : None
xlrd : None
xlwt : None
numba : 0.50.1

[jorisvandenbossche]

cc @mroeschke

[goodwanghan]

I had the same issue on groupby. To reproduce the issue:

import pandas as pd
from pandas.core.window.indexers import BaseIndexer

df = pd.DataFrame([[0, 1],[1, 20]], columns=["a", "b"])
# this should raise NotImplementedError (works in 1.0.2)
df.groupby(["a"]).rolling(window=BaseIndexer(),min_periods=1).min()

and count is an exception, it works with grouby rolling, this will raise:

df.groupby(["a"]).rolling(window=BaseIndexer(),min_periods=0).count()

[TomAugspurger]

FYI @goodwanghan, comments like "Please fix" come off as rude.

[goodwanghan]

FYI @goodwanghan, comments like "Please fix" come off as rude.

Thank you, and I feel very sorry about that. I have changed.