refactor: comply to more ruff and pyright rules

.pre-commit-config.yaml

@@ -11,17 +11,10 @@ repos:
     hooks:
     -   id: isort
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.13.2
+    rev: v0.13.3
     hooks:
     -   id: ruff-check
-        args: [
-          --exit-non-zero-on-fix,
-          --target-version, py39,
-          --extend-select, "PYI,UP,RUF100",
-          --ignore, "E501,E731,F841,PYI042",
-          --per-file-ignores, "_*.pyi:PYI001",
-          --fix
-        ]
+        args: [--exit-non-zero-on-fix]
 -   repo: https://github.com/codespell-project/codespell
     rev: v2.4.1
     hooks:

pandas-stubs/core/generic.pyi

@@ -29,6 +29,7 @@ from pandas.core.series import (
 import sqlalchemy.engine
 from typing_extensions import (
     Concatenate,
+    Never,
     Self,
 )
 
@@ -315,6 +316,42 @@ class NDFrame(indexing.IndexingMixin):
     @final
     def __delitem__(self, idx: Hashable) -> None: ...
     @overload
+    def drop(
+        self,
+        labels=...,
+        *,
+        axis=...,
+        index: None,
+        columns=...,
+        level=...,
+        inplace=...,
+        errors=...,
+    ) -> Never: ...
+    @overload
+    def drop(
+        self,
+        labels=...,
+        *,
+        axis=...,
+        index=...,
+        columns: None,
+        level=...,
+        inplace=...,
+        errors=...,
+    ) -> Never: ...
+    @overload
+    def drop(
+        self,
+        labels: None,
+        *,
+        axis=...,
+        index=...,
+        columns=...,
+        level=...,
+        inplace=...,
+        errors=...,
+    ) -> Never: ...
+    @overload
     def drop(
         self,
         labels: None = ...,

pyproject.toml

@@ -138,7 +138,7 @@ args = [
 
 
 [tool.black]
-target-version = ['py310']
+target-version = ['py39']
 
 [tool.isort]
 known_pre_libs = "pandas._config"
@@ -168,6 +168,25 @@ force_grid_wrap = 2
 force_sort_within_sections = true
 skip_glob = "env"
 
+
+[tool.ruff]
+target-version = "py39"
+fix = true
+
+
+[tool.ruff.lint]
+extend-select = ["PYI", "UP", "RUF100"]
+ignore = [
+  "E501",   # https://docs.astral.sh/ruff/rules/line-too-long/
+  "E731",   # https://docs.astral.sh/ruff/rules/lambda-assignment/
+  "PYI042", # https://docs.astral.sh/ruff/rules/snake-case-type-alias/
+]
+
+
+[tool.ruff.lint.per-file-ignores]
+"_*.pyi" = ["PYI001"]
+
+
 [tool.mypy]
 # Import discovery
 namespace_packages = false

tests/indexes/test_indexes.py

@@ -1379,7 +1379,7 @@ def test_index_categorical() -> None:
 def test_disallow_empty_index() -> None:
     # From GH 826
     if TYPE_CHECKING_INVALID_USAGE:
-        i0 = pd.Index()  # type: ignore[call-overload] # pyright: ignore[reportCallIssue]
+        _0 = pd.Index()  # type: ignore[call-overload] # pyright: ignore[reportCallIssue]
 
 
 def test_datetime_index_max_min_reductions() -> None:

tests/scalars/test_scalars.py

@@ -2109,5 +2109,5 @@ def test_period_methods() -> None:
 
 def test_nattype_hashable() -> None:
     # GH 827
-    aset = {pd.NaT}
+    _aset = {pd.NaT}
     check(assert_type(pd.NaT.__hash__(), int), int)

tests/series/test_series.py

@@ -310,16 +310,20 @@ def test_types_drop() -> None:
 
 def test_arguments_drop() -> None:
     # GH 950
+    s = pd.Series([0, 1, 2])
     if TYPE_CHECKING_INVALID_USAGE:
-        s = pd.Series([0, 1, 2])
-        res1 = s.drop()  # type: ignore[call-overload] # pyright: ignore[reportCallIssue]
-        res2 = s.drop([0], columns=["col1"])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
-        res3 = s.drop([0], index=[0])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
-        # These should also fail, but `None` is Hasheable and i do not know how
-        # to type hint a non-None hashable.
-        # res4 = s.drop(columns=None)
-        # res5 = s.drop(index=None)
-        # res6 = s.drop(None)
+        _res1 = s.drop()  # type: ignore[call-overload] # pyright: ignore[reportCallIssue]
+        _res2 = s.drop([0], columns=["col1"])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
+        _res3 = s.drop([0], index=[0])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
+
+    def _never_checker0() -> None:  # pyright: ignore[reportUnusedFunction]
+        assert_type(s.drop(columns=None), Never)
+
+    def _never_checker1() -> None:  # pyright: ignore[reportUnusedFunction]
+        assert_type(s.drop(index=None), Never)
+
+    def _never_checker2() -> None:  # pyright: ignore[reportUnusedFunction]
+        assert_type(s.drop(None), Never)
 
 
 def test_types_drop_multilevel() -> None:
@@ -817,14 +821,14 @@ def test_types_element_wise_arithmetic() -> None:
         assert_type(s.div(s2, fill_value=0), "pd.Series[float]"), pd.Series, np.float64
     )
 
-    res_floordiv: pd.Series = s // s2
-    res_floordiv2: pd.Series = s.floordiv(s2, fill_value=0)
+    _res_floordiv: pd.Series = s // s2
+    _res_floordiv2: pd.Series = s.floordiv(s2, fill_value=0)
 
-    res_mod: pd.Series = s % s2
-    res_mod2: pd.Series = s.mod(s2, fill_value=0)
+    _res_mod: pd.Series = s % s2
+    _res_mod2: pd.Series = s.mod(s2, fill_value=0)
 
-    res_pow: pd.Series = s ** s2.abs()
-    res_pow2: pd.Series = s.pow(s2.abs(), fill_value=0)
+    _res_pow: pd.Series = s ** s2.abs()
+    _res_pow2: pd.Series = s.pow(s2.abs(), fill_value=0)
 
     check(assert_type(divmod(s, s2), tuple["pd.Series[int]", "pd.Series[int]"]), tuple)
 
@@ -847,16 +851,16 @@ def test_types_scalar_arithmetic() -> None:
         assert_type(s.div(2, fill_value=0), "pd.Series[float]"), pd.Series, np.floating
     )
 
-    res_floordiv: pd.Series = s // 2
-    res_floordiv2: pd.Series = s.floordiv(2, fill_value=0)
+    _res_floordiv: pd.Series = s // 2
+    _res_floordiv2: pd.Series = s.floordiv(2, fill_value=0)
 
-    res_mod: pd.Series = s % 2
-    res_mod2: pd.Series = s.mod(2, fill_value=0)
+    _res_mod: pd.Series = s % 2
+    _res_mod2: pd.Series = s.mod(2, fill_value=0)
 
-    res_pow: pd.Series = s**2
-    res_pow1: pd.Series = s**0
-    res_pow2: pd.Series = s**0.213
-    res_pow3: pd.Series = s.pow(0.5)
+    _res_pow: pd.Series = s**2
+    _res_pow1: pd.Series = s**0
+    _res_pow2: pd.Series = s**0.213
+    _res_pow3: pd.Series = s.pow(0.5)
 
 
 def test_types_groupby() -> None:
@@ -1478,7 +1482,7 @@ def add1(x: int) -> int:
     )
 
     if TYPE_CHECKING_INVALID_USAGE:
-        s7 = pd.Series([1, 2, 3]).rename({1: [3, 4, 5]})  # type: ignore[dict-item] # pyright: ignore[reportArgumentType]
+        _s7 = pd.Series([1, 2, 3]).rename({1: [3, 4, 5]})  # type: ignore[dict-item] # pyright: ignore[reportArgumentType]
 
 
 def test_types_ne() -> None:
@@ -1606,7 +1610,7 @@ def test_series_multiindex_getitem() -> None:
     s = pd.Series(
         [1, 2, 3, 4], index=pd.MultiIndex.from_product([["a", "b"], ["x", "y"]])
     )
-    s1: pd.Series = s["a", :]
+    _s1: pd.Series = s["a", :]
 
 
 def test_reset_index() -> None:
@@ -1800,7 +1804,7 @@ def test_iloc_setitem_ndarray() -> None:
 
 def test_types_iter() -> None:
     s = pd.Series([1, 2, 3], dtype=int)
-    iterable: Iterable[int] = s
+    _iterable: Iterable[int] = s
     check(assert_type(iter(s), Iterator[int]), Iterator, int)
     check(assert_type(next(iter(s)), int), int)
 
@@ -3114,7 +3118,7 @@ def test_to_json_mode() -> None:
     check(assert_type(result2, str), str)
     check(assert_type(result4, str), str)
     if TYPE_CHECKING_INVALID_USAGE:
-        result3 = s.to_json(orient="records", lines=False, mode="a")  # type: ignore[call-overload] # pyright: ignore[reportArgumentType,reportCallIssue]
+        _result3 = s.to_json(orient="records", lines=False, mode="a")  # type: ignore[call-overload] # pyright: ignore[reportArgumentType,reportCallIssue]
 
 
 def test_groupby_diff() -> None:
@@ -3730,7 +3734,7 @@ def test_series_bool_fails() -> None:
         # mypy doesn't seem to figure that out, but pyright does
         if s == "foo":  # pyright: ignore[reportGeneralTypeIssues]
             # Next line is unreachable.
-            a = s[0]
+            _a = s[0]
             assert False
     except ValueError:
         pass
@@ -3905,7 +3909,7 @@ def test_series_index_type() -> None:
     )
 
     if TYPE_CHECKING_INVALID_USAGE:
-        t = pd.Series([1, 2], index="ab")  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
+        _t = pd.Series([1, 2], index="ab")  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
 
 
 def test_timedelta_index_cumprod() -> None:

tests/test_frame.py

@@ -177,22 +177,22 @@ def test_types_append() -> None:
     df = pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]})
     df2 = pd.DataFrame({"col1": [10, 20], "col2": [30, 40]})
     if TYPE_CHECKING_INVALID_USAGE:
-        res1: pd.DataFrame = df.append(df2)  # type: ignore[operator] # pyright: ignore[reportCallIssue]
-        res2: pd.DataFrame = df.append([1, 2, 3])  # type: ignore[operator] # pyright: ignore[reportCallIssue]
-        res3: pd.DataFrame = df.append([[1, 2, 3]])  # type: ignore[operator] # pyright: ignore[reportCallIssue]
-        res4: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res1: pd.DataFrame = df.append(df2)  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res2: pd.DataFrame = df.append([1, 2, 3])  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res3: pd.DataFrame = df.append([[1, 2, 3]])  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res4: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
             {("a", 1): [1, 2, 3], "b": df2}, ignore_index=True
         )
-        res5: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res5: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
             {1: [1, 2, 3]}, ignore_index=True
         )
-        res6: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res6: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
             {1: [1, 2, 3], "col2": [1, 2, 3]}, ignore_index=True
         )
-        res7: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res7: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
             pd.Series([5, 6]), ignore_index=True
         )
-        res8: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
+        _res8: pd.DataFrame = df.append(  # type: ignore[operator] # pyright: ignore[reportCallIssue]
             pd.Series([5, 6], index=["col1", "col2"]), ignore_index=True
         )
 
@@ -478,16 +478,20 @@ def test_types_drop() -> None:
 
 def test_arguments_drop() -> None:
     # GH 950
+    df = pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]})
     if TYPE_CHECKING_INVALID_USAGE:
-        df = pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]})
-        res1 = df.drop()  # type: ignore[call-overload] # pyright: ignore[reportCallIssue]
-        res2 = df.drop([0], columns=["col1"])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
-        res3 = df.drop([0], index=[0])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
-        # These should also fail, but `None` is Hasheable and i do not know how
-        # to type hint a non-None hashable.
-        # res4 = df.drop(columns=None)
-        # res5 = df.drop(index=None)
-        # res6 = df.drop(None)
+        _res1 = df.drop()  # type: ignore[call-overload] # pyright: ignore[reportCallIssue]
+        _res2 = df.drop([0], columns=["col1"])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
+        _res3 = df.drop([0], index=[0])  # type: ignore[call-overload] # pyright: ignore[reportCallIssue, reportArgumentType]
+
+    def _never_checker0() -> None:  # pyright: ignore[reportUnusedFunction]
+        assert_type(df.drop(columns=None), Never)
+
+    def _never_checker1() -> None:  # pyright: ignore[reportUnusedFunction]
+        assert_type(df.drop(index=None), Never)
+
+    def _never_checker2() -> None:  # pyright: ignore[reportUnusedFunction]
+        assert_type(df.drop(None), Never)
 
 
 def test_types_dropna() -> None:
@@ -4404,7 +4408,7 @@ def test_to_json_mode() -> None:
     check(assert_type(result2, str), str)
     check(assert_type(result4, str), str)
     if TYPE_CHECKING_INVALID_USAGE:
-        result3 = df.to_json(orient="records", lines=False, mode="a")  # type: ignore[call-overload] # pyright: ignore[reportArgumentType,reportCallIssue]
+        _result3 = df.to_json(orient="records", lines=False, mode="a")  # type: ignore[call-overload] # pyright: ignore[reportArgumentType,reportCallIssue]
 
 
 def test_interpolate_inplace() -> None:
@@ -4547,7 +4551,7 @@ def test_frame_bool_fails() -> None:
         # mypy doesn't seem to figure that out, but pyright does
         if df == "foo":  # pyright: ignore[reportGeneralTypeIssues]
             # Next line is unreachable.
-            s = df["a"]
+            _s = df["a"]
     except ValueError:
         pass
 

tests/test_string_accessors.py

@@ -280,7 +280,7 @@ def test_string_accessors_list_series():
     # rsplit doesn't accept compiled pattern
     # it doesn't raise at runtime but produces a nan
     if TYPE_CHECKING_INVALID_USAGE:
-        bad_rsplit_result = s.str.rsplit(
+        _bad_rsplit_result = s.str.rsplit(
             re.compile(r"a")  # type: ignore[call-overload] # pyright: ignore[reportArgumentType]
         )
 
@@ -300,7 +300,7 @@ def test_string_accessors_list_index():
     # rsplit doesn't accept compiled pattern
     # it doesn't raise at runtime but produces a nan
     if TYPE_CHECKING_INVALID_USAGE:
-        bad_rsplit_result = idx.str.rsplit(
+        _bad_rsplit_result = idx.str.rsplit(
             re.compile(r"a")  # type: ignore[call-overload] # pyright: ignore[reportArgumentType]
         )
 

tests/test_timefuncs.py

@@ -293,9 +293,9 @@ def test_fail_on_adding_two_timestamps() -> None:
     s1 = pd.Series(pd.to_datetime(["2022-05-01", "2022-06-01"]))
     s2 = pd.Series(pd.to_datetime(["2022-05-15", "2022-06-15"]))
     if TYPE_CHECKING_INVALID_USAGE:
-        ssum = s1 + s2  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
+        _ssum = s1 + s2  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
         ts = pd.Timestamp("2022-06-30")
-        tsum = s1 + ts  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
+        _tsum = s1 + ts  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
 
 
 def test_dtindex_tzinfo() -> None:
@@ -1706,9 +1706,9 @@ def test_timedelta64_and_arithmatic_operator() -> None:
     td = np.timedelta64(1, "D")
     check(assert_type((s3 / td), "pd.Series[float]"), pd.Series, float)
     if TYPE_CHECKING_INVALID_USAGE:
-        r1 = s1 * td  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
-        r2 = s1 / td  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
-        r3 = s3 * td  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
+        _1 = s1 * td  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
+        _2 = s1 / td  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
+        _3 = s3 * td  # type: ignore[operator] # pyright: ignore[reportOperatorIssue]
 
 
 def test_timedeltaseries_add_timestampseries() -> None: