BUG: wrong df.groupby().groups when grouping with [Grouper(freq=), ...] (GH33132)

[falcaopetri]

• closes BUG: wrong df.groupby().groups when grouping with [Grouper(freq=), ...] #33132
• tests added / passed
• passes black pandas
• passes git diff upstream/master -u -- "*.py" | flake8 --diff
• whatsnew entry

[falcaopetri]

The proposed change breaks the equivalence of df.groupby(grouper) and df.groupby([grouper]), captured by test

pandas/pandas/tests/groupby/test_timegrouper.py

Line 740 in 8415f9b

def test_scalar_call_versus_list_call(self):

Currently executing this test results in:
(The correct behavior is the one in result, and not in expected.)

import pandas as pd
import pandas._testing as tm

data_frame = {
    "location": ["shanghai", "beijing", "shanghai"],
    "time": pd.Series(
        ["2017-08-09 13:32:23", "2017-08-11 23:23:15", "2017-08-11 22:23:15"],
        dtype="datetime64[ns]",
    ),
    "value": [1, 2, 3],
}
data_frame = pd.DataFrame(data_frame).set_index("time")
grouper = pd.Grouper(freq="D")

grouped = data_frame.groupby(grouper)
result = grouped.count()
grouped = data_frame.groupby([grouper])
expected = grouped.count()

print(result)
#             location  value
# time                       
# 2017-08-09         1      1
# 2017-08-10         0      0
# 2017-08-11         2      2
print(expected)
#             location  value
# time                       
# 2017-08-09         1      1
# 2017-08-11         2      2

The problem occurs when we have something like df.groupby([pd.Grouper(freq=)]) (i.e., a list with one element of type TimeGrouper).

One way to fix this is by capturing the keys arg in

pandas/pandas/core/groupby/groupby.py

Lines 356 to 363 in 8305a85

	class _GroupBy(PandasObject, SelectionMixin):
	_group_selection = None
	_apply_whitelist: FrozenSet[str] = frozenset()
	def __init__(
	self,
	obj: NDFrame,
	keys: Optional[_KeysArgType] = None,

and, if it is a [pd.Grouper], we transform it to a pd.Grouper.

Does this modification make sense?

[jreback]

please merge master and see if you can resolve the issues.

[mroeschke]

Closing this PR as stale, but feel free to ping if you'd like to continue working on it.