TYP: resample

pandas/core/resample.py

@@ -4,6 +4,7 @@
 from datetime import timedelta
 from textwrap import dedent
 from typing import (
+    TYPE_CHECKING,
     Callable,
     no_type_check,
 )
@@ -12,6 +13,7 @@
 
 from pandas._libs import lib
 from pandas._libs.tslibs import (
+    BaseOffset,
     IncompatibleFrequency,
     NaT,
     Period,
@@ -84,6 +86,9 @@
     Tick,
 )
 
+if TYPE_CHECKING:
+    from typing import Literal
+
 _shared_docs_kwargs: dict[str, str] = {}
 
 
@@ -489,11 +494,11 @@ def _apply_loffset(self, result):
         self.loffset = None
         return result
 
-    def _get_resampler_for_grouping(self, groupby, **kwargs):
+    def _get_resampler_for_grouping(self, groupby):
         """
         Return the correct class for resampling with groupby.
         """
-        return self._resampler_for_grouping(self, groupby=groupby, **kwargs)
+        return self._resampler_for_grouping(self, groupby=groupby)
 
     def _wrap_result(self, result):
         """
@@ -1039,9 +1044,10 @@ class _GroupByMixin(PandasObject):
     Provide the groupby facilities.
     """
 
-    _attributes: list[str]
+    _attributes: list[str]  # in practice the same as Resampler._attributes
 
-    def __init__(self, obj, *args, **kwargs):
+    def __init__(self, obj, **kwargs):
+        # reached via ._gotitem and _get_resampler_for_grouping
 
         parent = kwargs.pop("parent", None)
         groupby = kwargs.pop("groupby", None)
@@ -1450,7 +1456,7 @@ class TimeGrouper(Grouper):
     def __init__(
         self,
         freq="Min",
-        closed: str | None = None,
+        closed: Literal["left", "right"] | None = None,
         label: str | None = None,
         how="mean",
         axis=0,
@@ -1822,8 +1828,13 @@ def _take_new_index(
 
 
 def _get_timestamp_range_edges(
-    first, last, freq, closed="left", origin="start_day", offset=None
-):
+    first: Timestamp,
+    last: Timestamp,
+    freq: BaseOffset,
+    closed: Literal["right", "left"] = "left",
+    origin="start_day",
+    offset: Timedelta | None = None,
+) -> tuple[Timestamp, Timestamp]:
     """
     Adjust the `first` Timestamp to the preceding Timestamp that resides on
     the provided offset. Adjust the `last` Timestamp to the following
@@ -1895,8 +1906,13 @@ def _get_timestamp_range_edges(
 
 
 def _get_period_range_edges(
-    first, last, freq, closed="left", origin="start_day", offset=None
-):
+    first: Period,
+    last: Period,
+    freq: BaseOffset,
+    closed: Literal["right", "left"] = "left",
+    origin="start_day",
+    offset: Timedelta | None = None,
+) -> tuple[Period, Period]:
     """
     Adjust the provided `first` and `last` Periods to the respective Period of
     the given offset that encompasses them.
@@ -1959,7 +1975,12 @@ def _insert_nat_bin(
 
 
 def _adjust_dates_anchored(
-    first, last, freq, closed="right", origin="start_day", offset=None
+    first,
+    last,
+    freq,
+    closed: Literal["right", "left"] = "right",
+    origin="start_day",
+    offset: Timedelta | None = None,
 ):
     # First and last offsets should be calculated from the start day to fix an
     # error cause by resampling across multiple days when a one day period is
@@ -2029,7 +2050,14 @@ def _adjust_dates_anchored(
     return fresult, lresult
 
 
-def asfreq(obj, freq, method=None, how=None, normalize=False, fill_value=None):
+def asfreq(
+    obj: FrameOrSeries,
+    freq,
+    method=None,
+    how=None,
+    normalize: bool = False,
+    fill_value=None,
+) -> FrameOrSeries:
     """
     Utility frequency conversion method for Series/DataFrame.
 

pandas/core/sorting.py

@@ -232,7 +232,7 @@ def decons_group_index(comp_labels, shape):
     return label_list[::-1]
 
 
-def decons_obs_group_ids(comp_ids, obs_ids, shape, labels, xnull: bool):
+def decons_obs_group_ids(comp_ids: np.ndarray, obs_ids, shape, labels, xnull: bool):
     """
     Reconstruct labels from observed group ids.
 
@@ -360,6 +360,10 @@ def nargsort(
     key : Optional[Callable], default None
     mask : Optional[np.ndarray], default None
         Passed when called by ExtensionArray.argsort.
+
+    Returns
+    -------
+    np.ndarray[np.intp]
     """
 
     if key is not None:
@@ -404,7 +408,7 @@ def nargsort(
         indexer = np.concatenate([nan_idx, indexer])
     else:
         raise ValueError(f"invalid na_position: {na_position}")
-    return indexer
+    return ensure_platform_int(indexer)
 
 
 def nargminmax(values, method: str, axis: int = 0):
@@ -644,7 +648,9 @@ def get_group_index_sorter(
     return ensure_platform_int(sorter)
 
 
-def compress_group_index(group_index, sort: bool = True):
+def compress_group_index(
+    group_index: np.ndarray, sort: bool = True
+) -> tuple[np.ndarray, np.ndarray]:
     """
     Group_index is offsets into cartesian product of all possible labels. This
     space can be huge, so this function compresses it, by computing offsets
@@ -682,7 +688,7 @@ def _reorder_by_uniques(
     sorter = uniques.argsort()
 
     # reverse_indexer is where elements came from
-    reverse_indexer = np.empty(len(sorter), dtype=np.int64)
+    reverse_indexer = np.empty(len(sorter), dtype=np.intp)
     reverse_indexer.put(sorter, np.arange(len(sorter)))
 
     mask = labels < 0

pandas/core/window/rolling.py

@@ -43,10 +43,7 @@
 )
 from pandas.core.dtypes.generic import (
     ABCDataFrame,
-    ABCDatetimeIndex,
-    ABCPeriodIndex,
     ABCSeries,
-    ABCTimedeltaIndex,
 )
 from pandas.core.dtypes.missing import notna
 
@@ -58,8 +55,11 @@
 )
 import pandas.core.common as com
 from pandas.core.indexes.api import (
+    DatetimeIndex,
     Index,
     MultiIndex,
+    PeriodIndex,
+    TimedeltaIndex,
 )
 from pandas.core.internals import ArrayManager
 from pandas.core.reshape.concat import concat
@@ -1455,9 +1455,7 @@ def validate(self):
         # we allow rolling on a datetimelike index
         if (
             self.obj.empty
-            or isinstance(
-                self._on, (ABCDatetimeIndex, ABCTimedeltaIndex, ABCPeriodIndex)
-            )
+            or isinstance(self._on, (DatetimeIndex, TimedeltaIndex, PeriodIndex))
         ) and isinstance(self.window, (str, BaseOffset, timedelta)):
 
             self._validate_monotonic()
@@ -1470,7 +1468,7 @@ def validate(self):
                     f"passed window {self.window} is not "
                     "compatible with a datetimelike index"
                 ) from err
-            if isinstance(self._on, ABCPeriodIndex):
+            if isinstance(self._on, PeriodIndex):
                 self._win_freq_i8 = freq.nanos / (self._on.freq.nanos / self._on.freq.n)
             else:
                 self._win_freq_i8 = freq.nanos