TYP: to_dict use MutableMapping (#796)

* TYP: to_dict use MutableMapping

* remove deprecation warning

* mypy

Author: twoertwein

pandas-stubs/core/frame.pyi

@@ -4,6 +4,7 @@ from collections.abc import (
     Iterable,
     Iterator,
     Mapping,
+    MutableMapping,
     Sequence,
 )
 import datetime
@@ -270,61 +271,67 @@ class DataFrame(NDFrame, OpsMixin):
         na_value: Scalar = ...,
     ) -> np.ndarray: ...
     @overload
-    def to_dict(
+    def to_dict(  # type: ignore[misc]
         self,
         orient: Literal["records"],
-        into: Mapping | type[Mapping],
+        *,
+        into: MutableMapping | type[MutableMapping],
         index: Literal[True] = ...,
-    ) -> list[Mapping[Hashable, Any]]: ...
+    ) -> list[MutableMapping[Hashable, Any]]: ...
     @overload
     def to_dict(
         self,
         orient: Literal["records"],
-        into: None = ...,
+        *,
+        into: type[dict] = ...,
         index: Literal[True] = ...,
     ) -> list[dict[Hashable, Any]]: ...
     @overload
-    def to_dict(
+    def to_dict(  # type: ignore[misc]
         self,
         orient: Literal["dict", "list", "series", "index"],
-        into: Mapping | type[Mapping],
+        *,
+        into: MutableMapping | type[MutableMapping],
         index: Literal[True] = ...,
-    ) -> Mapping[Hashable, Any]: ...
+    ) -> MutableMapping[Hashable, Any]: ...
     @overload
-    def to_dict(
+    def to_dict(  # type: ignore[misc]
         self,
         orient: Literal["split", "tight"],
-        into: Mapping | type[Mapping],
+        *,
+        into: MutableMapping | type[MutableMapping],
         index: bool = ...,
-    ) -> Mapping[Hashable, Any]: ...
+    ) -> MutableMapping[Hashable, Any]: ...
     @overload
-    def to_dict(
+    def to_dict(  # type: ignore[misc]
         self,
         orient: Literal["dict", "list", "series", "index"] = ...,
         *,
-        into: Mapping | type[Mapping],
+        into: MutableMapping | type[MutableMapping],
         index: Literal[True] = ...,
-    ) -> Mapping[Hashable, Any]: ...
+    ) -> MutableMapping[Hashable, Any]: ...
     @overload
-    def to_dict(
+    def to_dict(  # type: ignore[misc]
         self,
         orient: Literal["split", "tight"] = ...,
         *,
-        into: Mapping | type[Mapping],
+        into: MutableMapping | type[MutableMapping],
         index: bool = ...,
-    ) -> Mapping[Hashable, Any]: ...
+    ) -> MutableMapping[Hashable, Any]: ...
     @overload
     def to_dict(
         self,
         orient: Literal["dict", "list", "series", "index"] = ...,
-        into: None = ...,
+        *,
+        into: type[dict] = ...,
         index: Literal[True] = ...,
     ) -> dict[Hashable, Any]: ...
     @overload
     def to_dict(
         self,
         orient: Literal["split", "tight"] = ...,
-        into: None = ...,
+        *,
+        into: type[dict] = ...,
         index: bool = ...,
     ) -> dict[Hashable, Any]: ...
     def to_gbq(

pandas-stubs/core/series.pyi

@@ -4,6 +4,7 @@ from collections.abc import (
     Iterable,
     Iterator,
     Mapping,
+    MutableMapping,
     Sequence,
 )
 from datetime import (
@@ -535,9 +536,11 @@ class Series(IndexOpsMixin[S1], NDFrame):
     def items(self) -> Iterable[tuple[Hashable, S1]]: ...
     def keys(self) -> list: ...
     @overload
-    def to_dict(self) -> dict[Any, S1]: ...
+    def to_dict(self, *, into: type[dict] = ...) -> dict[Any, S1]: ...
     @overload
-    def to_dict(self, into: type[Mapping] | Mapping) -> Mapping[Hashable, S1]: ...
+    def to_dict(
+        self, *, into: type[MutableMapping] | MutableMapping
+    ) -> MutableMapping[Hashable, S1]: ...
     def to_frame(self, name: object | None = ...) -> DataFrame: ...
     @overload
     def groupby(

pyproject.toml

@@ -35,12 +35,11 @@ types-pytz = ">= 2022.1.1"
 numpy = { version = ">=1.26.0", python = "<3.13" }
 
 [tool.poetry.dev-dependencies]
-mypy = "1.5.1"
+mypy = "1.6.0"
 pandas = "2.1.1"
 pyarrow = ">=10.0.1"
 pytest = ">=7.1.2"
-# pyright 1.1.329 has bug fixed in 1.1.330
-pyright = "1.1.328"
+pyright = ">=1.1.331"
 poethepoet = ">=0.16.5"
 loguru = ">=0.6.0"
 typing-extensions = ">=4.4.0"

tests/test_frame.py

@@ -6,6 +6,7 @@
     Iterable,
     Iterator,
     Mapping,
+    MutableMapping,
 )
 import csv
 import datetime
@@ -2318,11 +2319,11 @@ def func() -> MyDataFrame[int]:
     func()
 
 
-def test_to_xarray():
+def test_to_xarray() -> None:
     check(assert_type(DF.to_xarray(), xr.Dataset), xr.Dataset)
 
 
-def test_to_records():
+def test_to_records() -> None:
     check(assert_type(DF.to_records(False, "int8"), np.recarray), np.recarray)
     check(
         assert_type(DF.to_records(False, index_dtypes=np.int8), np.recarray),
@@ -2336,23 +2337,33 @@ def test_to_records():
     )
 
 
-def test_to_dict():
+def test_to_dict() -> None:
     check(assert_type(DF.to_dict(), dict[Hashable, Any]), dict)
     check(assert_type(DF.to_dict("split"), dict[Hashable, Any]), dict)
 
-    target: Mapping = defaultdict(list)
-    check(assert_type(DF.to_dict(into=target), Mapping[Hashable, Any]), defaultdict)
+    target: MutableMapping = defaultdict(list)
+    check(
+        assert_type(DF.to_dict(into=target), MutableMapping[Hashable, Any]), defaultdict
+    )
     target = defaultdict(list)
     check(
-        assert_type(DF.to_dict("tight", into=target), Mapping[Hashable, Any]),
+        assert_type(DF.to_dict("tight", into=target), MutableMapping[Hashable, Any]),
         defaultdict,
     )
     target = defaultdict(list)
     check(assert_type(DF.to_dict("records"), list[dict[Hashable, Any]]), list)
     check(
-        assert_type(DF.to_dict("records", into=target), list[Mapping[Hashable, Any]]),
+        assert_type(
+            DF.to_dict("records", into=target), list[MutableMapping[Hashable, Any]]
+        ),
         list,
     )
+    if TYPE_CHECKING_INVALID_USAGE:
+
+        def test(mapping: Mapping) -> None:  # pyright: ignore[reportUnusedFunction]
+            DF.to_dict(  # type: ignore[call-overload]
+                into=mapping  # pyright: ignore[reportGeneralTypeIssues]
+            )
 
 
 def test_neg() -> None:

tests/test_io.py

@@ -360,7 +360,7 @@ def test_hdf():
         check(assert_type(read_hdf(path), Union[DataFrame, Series]), DataFrame)
 
 
-def test_hdfstore():
+def test_hdfstore() -> None:
     with ensure_clean() as path:
         store = HDFStore(path, model="w")
         check(assert_type(store, HDFStore), HDFStore)
@@ -372,91 +372,67 @@ def test_hdfstore():
             assert_type(store.select("df", start=0, stop=1), Union[DataFrame, Series]),
             DataFrame,
         )
-        with pytest_warns_bounded(
-            DeprecationWarning,
-            match="`alltrue` is deprecated as of NumPy 1.25.0",
-            lower="1.24.99",
-            version_str=np.__version__,
-        ):
-            check(
-                assert_type(
-                    store.select("df", where="index>=1"), Union[DataFrame, Series]
-                ),
-                DataFrame,
-            )
-            check(
-                assert_type(
-                    store.select("df", where=Term("index>=1")),
-                    Union[DataFrame, Series],
-                ),
-                DataFrame,
-            )
-            check(
-                assert_type(
-                    store.select("df", where=[Term("index>=1")]),
-                    Union[DataFrame, Series],
-                ),
-                DataFrame,
-            )
-            check(assert_type(store.get("df"), Union[DataFrame, Series]), DataFrame)
-            for key in store:
-                check(assert_type(key, str), str)
-            check(assert_type(store.close(), None), type(None))
+        check(
+            assert_type(store.select("df", where="index>=1"), Union[DataFrame, Series]),
+            DataFrame,
+        )
+        check(
+            assert_type(
+                store.select("df", where=Term("index>=1")),
+                Union[DataFrame, Series],
+            ),
+            DataFrame,
+        )
+        check(
+            assert_type(
+                store.select("df", where=[Term("index>=1")]),
+                Union[DataFrame, Series],
+            ),
+            DataFrame,
+        )
+        check(assert_type(store.get("df"), Union[DataFrame, Series]), DataFrame)
+        for key in store:
+            check(assert_type(key, str), str)
+        check(assert_type(store.close(), None), type(None))
 
-            store = HDFStore(path, model="r")
-            check(
-                assert_type(read_hdf(store, "df"), Union[DataFrame, Series]),
-                DataFrame,
-            )
-            store.close()
+        store = HDFStore(path, model="r")
+        check(
+            assert_type(read_hdf(store, "df"), Union[DataFrame, Series]),
+            DataFrame,
+        )
+        store.close()
 
 
-def test_read_hdf_iterator():
-    with pytest_warns_bounded(
-        DeprecationWarning,
-        match="`alltrue` is deprecated as of NumPy 1.25.0",
-        lower="1.24.99",
-        version_str=np.__version__,
-    ):
-        with ensure_clean() as path:
-            with pytest_warns_bounded(
-                FutureWarning,
-                r".*all arguments of to_hdf except for the argument 'path_or_buf' will be keyword-only",
-                lower="2.1.99",
-            ):
-                check(
-                    assert_type(DF.to_hdf(path, "df", format="table"), None), type(None)
-                )
-            ti = read_hdf(path, chunksize=1)
-            check(assert_type(ti, TableIterator), TableIterator)
-            ti.close()
+def test_read_hdf_iterator() -> None:
+    with ensure_clean() as path:
+        with pytest_warns_bounded(
+            FutureWarning,
+            r".*all arguments of to_hdf except for the argument 'path_or_buf' will be keyword-only",
+            lower="2.1.99",
+        ):
+            check(assert_type(DF.to_hdf(path, "df", format="table"), None), type(None))
+        ti = read_hdf(path, chunksize=1)
+        check(assert_type(ti, TableIterator), TableIterator)
+        ti.close()
 
-            ti = read_hdf(path, "df", iterator=True)
-            check(assert_type(ti, TableIterator), TableIterator)
-            for _ in ti:
-                pass
-            ti.close()
+        ti = read_hdf(path, "df", iterator=True)
+        check(assert_type(ti, TableIterator), TableIterator)
+        for _ in ti:
+            pass
+        ti.close()
 
 
-def test_hdf_context_manager():
-    with pytest_warns_bounded(
-        DeprecationWarning,
-        match="`alltrue` is deprecated as of NumPy 1.25.0",
-        lower="1.24.99",
-        version_str=np.__version__,
-    ):
-        with ensure_clean() as path:
-            with pytest_warns_bounded(
-                FutureWarning,
-                r".*all arguments of to_hdf except for the argument 'path_or_buf' will be keyword-only",
-                lower="2.1.99",
-            ):
-                check(
-                    assert_type(DF.to_hdf(path, "df", format="table"), None), type(None)
-                )
-            with HDFStore(path, mode="r") as store:
-                check(assert_type(store.is_open, bool), bool)
-                check(assert_type(store.get("df"), Union[DataFrame, Series]), DataFrame)
+def test_hdf_context_manager() -> None:
+    with ensure_clean() as path:
+        with pytest_warns_bounded(
+            FutureWarning,
+            r".*all arguments of to_hdf except for the argument 'path_or_buf' will be keyword-only",
+            lower="2.1.99",
+        ):
+            check(assert_type(DF.to_hdf(path, "df", format="table"), None), type(None))
+        with HDFStore(path, mode="r") as store:
+            check(assert_type(store.is_open, bool), bool)
+            check(assert_type(store.get("df"), Union[DataFrame, Series]), DataFrame)
 
 
 def test_hdf_series():