GH1055 Add pandas.api.typing to pandas-stubs (#1058)

* GH1055 Add pandas.api.typing to pandas-stubs

* GH1055 Fix lint

* GH1055 Fix assert_type

* GH1055 Fix tests

* Revert to generic

* GH1055 Fix tests

* GH1055 new test format

* GH1055 PR Feedback

* GH1055 TypeAlias Feedback

Author: loicdiridollou

pandas-stubs/api/typing/__init__.pyi

@@ -0,0 +1,29 @@
+from pandas.core.groupby import (
+    DataFrameGroupBy as DataFrameGroupBy,
+    SeriesGroupBy as SeriesGroupBy,
+)
+from pandas.core.resample import (
+    DatetimeIndexResamplerGroupby as DatetimeIndexResamplerGroupby,
+    PeriodIndexResamplerGroupby as PeriodIndexResamplerGroupby,
+    Resampler as Resampler,
+    TimedeltaIndexResamplerGroupby as TimedeltaIndexResamplerGroupby,
+    TimeGrouper as TimeGrouper,
+)
+from pandas.core.window import (
+    Expanding as Expanding,
+    ExpandingGroupby as ExpandingGroupby,
+    ExponentialMovingWindow as ExponentialMovingWindow,
+    ExponentialMovingWindowGroupby as ExponentialMovingWindowGroupby,
+    Rolling as Rolling,
+    RollingGroupby as RollingGroupby,
+    Window as Window,
+)
+
+from pandas._libs import NaTType as NaTType
+from pandas._libs.missing import NAType as NAType
+
+from pandas.io.json._json import JsonReader as JsonReader
+
+# SASReader is not defined so commenting it out for now
+# from pandas.io.sas.sasreader import SASReader as SASReader
+from pandas.io.stata import StataReader as StataReader

tests/test_api_typing.py

@@ -0,0 +1,218 @@
+"""Test module for classes in pandas.api.typing."""
+
+import numpy as np
+import pandas as pd
+from pandas._testing import ensure_clean
+from pandas.api.typing import (
+    DataFrameGroupBy,
+    DatetimeIndexResamplerGroupby,
+    Expanding,
+    ExpandingGroupby,
+    ExponentialMovingWindow,
+    ExponentialMovingWindowGroupby,
+    JsonReader,
+    NaTType,
+    NAType,
+    PeriodIndexResamplerGroupby,
+    Resampler,
+    Rolling,
+    RollingGroupby,
+    SeriesGroupBy,
+    StataReader,
+    TimedeltaIndexResamplerGroupby,
+    TimeGrouper,
+    Window,
+)
+import pytest
+from typing_extensions import (
+    TypeAlias,
+    assert_type,
+)
+
+from tests import check
+
+from pandas.io.json._json import read_json
+
+ResamplerGroupBy: TypeAlias = (
+    DatetimeIndexResamplerGroupby
+    | PeriodIndexResamplerGroupby
+    | TimedeltaIndexResamplerGroupby
+)
+
+
+def test_dataframegroupby():
+    df = pd.DataFrame({"a": [1, 2, 3]})
+    group = df.groupby("a")
+
+    def f1(gb: DataFrameGroupBy):
+        check(gb, DataFrameGroupBy)
+
+    f1(group)
+
+
+def test_seriesgroupby():
+    sr = pd.Series([1, 2, 3], index=pd.Index(["a", "b", "a"]))
+
+    def f1(gb: SeriesGroupBy):
+        check(gb, SeriesGroupBy)
+
+    f1(sr.groupby(level=0))
+
+
+def tests_datetimeindexersamplergroupby() -> None:
+    idx = pd.date_range("1999-1-1", periods=365, freq="D")
+    df = pd.DataFrame(
+        np.random.standard_normal((365, 2)), index=idx, columns=["col1", "col2"]
+    )
+    gb_df = df.groupby("col2")
+
+    def f1(gb: ResamplerGroupBy):
+        check(gb, DatetimeIndexResamplerGroupby)
+
+    f1(gb_df.resample("ME"))
+
+
+def test_timedeltaindexresamplergroupby() -> None:
+    idx = pd.TimedeltaIndex(["0 days", "1 days", "2 days", "3 days", "4 days"])
+    df = pd.DataFrame(
+        np.random.standard_normal((5, 2)), index=idx, columns=["col1", "col2"]
+    )
+    gb_df = df.groupby("col2")
+
+    def f1(gb: ResamplerGroupBy):
+        check(gb, TimedeltaIndexResamplerGroupby)
+
+    f1(gb_df.resample("1D"))
+
+
+@pytest.mark.skip("Resampling with a PeriodIndex is deprecated.")
+def test_periodindexresamplergroupby() -> None:
+    idx = pd.period_range("2020-01-28 09:00", periods=4, freq="D")
+    df = pd.DataFrame(data=4 * [range(2)], index=idx, columns=["a", "b"])
+
+    def f1(gb: ResamplerGroupBy):
+        check(gb, PeriodIndexResamplerGroupby)
+
+    f1(df.groupby("a").resample("3min"))
+
+
+def test_natype() -> None:
+    i64dt = pd.Int64Dtype()
+    check(assert_type(i64dt.na_value, NAType), NAType)
+
+
+def test_nattype() -> None:
+    td = pd.Timedelta("1 day")
+    as_nat = pd.NaT
+
+    check(assert_type(td + as_nat, NaTType), NaTType)
+
+
+def test_expanding() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: Expanding):
+        check(gb, Expanding)
+
+    f1(df.expanding())
+
+
+def test_expanding_groubpy() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: ExpandingGroupby):
+        check(gb, ExpandingGroupby)
+
+    f1(df.groupby("B").expanding())
+
+
+def test_ewm() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: ExponentialMovingWindow):
+        check(gb, ExponentialMovingWindow)
+
+    f1(df.ewm(2))
+
+
+def test_ewm_groubpy() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: ExponentialMovingWindowGroupby):
+        check(gb, ExponentialMovingWindowGroupby)
+
+    f1(df.groupby("B").ewm(2))
+
+
+def test_json_reader() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: JsonReader):
+        check(gb, JsonReader)
+
+    with ensure_clean() as path:
+        check(assert_type(df.to_json(path), None), type(None))
+        json_reader = read_json(path, chunksize=1, lines=True)
+        f1(json_reader)
+        json_reader.close()
+
+
+def test_resampler() -> None:
+    s = pd.Series([1, 2, 3, 4, 5], index=pd.date_range("20130101", periods=5, freq="s"))
+
+    def f1(gb: Resampler):
+        check(gb, Resampler)
+
+    f1(s.resample("3min"))
+
+
+def test_rolling() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: Rolling):
+        check(gb, Rolling)
+
+    f1(df.rolling(2))
+
+
+def test_rolling_groupby() -> None:
+    df = pd.DataFrame({"B": [0, 1, 2, np.nan, 4]})
+
+    def f1(gb: RollingGroupby):
+        check(gb, RollingGroupby)
+
+    f1(df.groupby("B").rolling(2))
+
+
+def test_timegrouper() -> None:
+    grouper = pd.Grouper(key="Publish date", freq="1W")
+
+    def f1(gb: TimeGrouper):
+        check(gb, TimeGrouper)
+
+    f1(grouper)
+
+
+def test_window() -> None:
+    ser = pd.Series([0, 1, 5, 2, 8])
+
+    def f1(gb: Window):
+        check(gb, Window)
+
+    f1(ser.rolling(2, win_type="gaussian"))
+
+
+def test_statereader() -> None:
+    df = pd.DataFrame([[1, 2], [3, 4]], columns=["col_1", "col_2"])
+    time_stamp = pd.Timestamp(2000, 2, 29, 14, 21)
+    variable_labels = {"col_1": "This is an example"}
+    with ensure_clean() as path:
+        df.to_stata(
+            path, time_stamp=time_stamp, variable_labels=variable_labels, version=None
+        )
+
+        def f1(gb: StataReader):
+            check(gb, StataReader)
+
+        with StataReader(path) as reader:
+            f1(reader)