Support numpy 2.0 in tests (#943)

* change np.float_ to np.double

* changes to support numpy 2.0

Author: Dr-Irv

pandas-stubs/_typing.pyi

@@ -115,7 +115,7 @@ BooleanDtypeArg: TypeAlias = (
     # Numpy bool type
     # https://numpy.org/doc/stable/reference/arrays.scalars.html#numpy.bool_
     | type[np.bool_]
-    | Literal["?", "b1", "bool8", "bool_"]
+    | Literal["?", "b1", "bool_"]
     # PyArrow boolean type and its string alias
     | Literal["bool[pyarrow]", "boolean[pyarrow]"]
 )
@@ -147,7 +147,7 @@ IntDtypeArg: TypeAlias = (
     | Literal["q", "longlong"]  # NOTE: int128 not assigned
     # https://numpy.org/doc/stable/reference/arrays.scalars.html#numpy.intp
     | type[np.intp]  # signed pointer (=`intptr_t`, platform dependent)
-    | Literal["p", "intp", "int0"]
+    | Literal["p", "intp"]
     # PyArrow integer types and their string aliases
     | Literal["int8[pyarrow]", "int16[pyarrow]", "int32[pyarrow]", "int64[pyarrow]"]
 )
@@ -176,7 +176,7 @@ UIntDtypeArg: TypeAlias = (
     | Literal["Q", "ulonglong"]  # NOTE: uint128 not assigned
     # https://numpy.org/doc/stable/reference/arrays.scalars.html#numpy.uintp
     | type[np.uintp]  # unsigned pointer (=`uintptr_t`, platform dependent)
-    | Literal["P", "uintp", "uint0"]
+    | Literal["P", "uintp"]
     # PyArrow unsigned integer types and their string aliases
     | Literal["uint8[pyarrow]", "uint16[pyarrow]", "uint32[pyarrow]", "uint64[pyarrow]"]
 )
@@ -361,7 +361,7 @@ BytesDtypeArg: TypeAlias = (
     # Numpy bytes type and its string alias
     # https://numpy.org/doc/stable/reference/arrays.scalars.html#numpy.bytes_
     | type[np.bytes_]
-    | Literal["S", "a", "bytes_", "bytes0", "string_"]
+    | Literal["S", "bytes_", "bytes0", "string_"]
     # PyArrow binary type and its string alias
     | Literal["binary[pyarrow]"]
 )

pandas-stubs/core/resample.pyi

@@ -163,7 +163,7 @@ class Resampler(BaseGroupBy[NDFrameT]):
     def count(self: Resampler[DataFrame]) -> DataFrame: ...
     def quantile(
         self,
-        q: float | list[float] | npt.NDArray[np.float_] | Series[float] = ...,
+        q: float | list[float] | npt.NDArray[np.double] | Series[float] = ...,
         **kwargs,
     ) -> NDFrameT: ...
 

pandas-stubs/core/reshape/tile.pyi

@@ -250,7 +250,7 @@ def qcut(
     retbins: Literal[True],
     precision: int = ...,
     duplicates: Literal["raise", "drop"] = ...,
-) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.float_]]: ...
+) -> tuple[npt.NDArray[np.intp], npt.NDArray[np.double]]: ...
 @overload
 def qcut(
     x: Series,
@@ -260,7 +260,7 @@ def qcut(
     retbins: Literal[True],
     precision: int = ...,
     duplicates: Literal["raise", "drop"] = ...,
-) -> tuple[Series, npt.NDArray[np.float_]]: ...
+) -> tuple[Series, npt.NDArray[np.double]]: ...
 @overload
 def qcut(
     x: Index | npt.NDArray | Sequence[int] | Sequence[float],
@@ -270,4 +270,4 @@ def qcut(
     retbins: Literal[True],
     precision: int = ...,
     duplicates: Literal["raise", "drop"] = ...,
-) -> tuple[Categorical, npt.NDArray[np.float_]]: ...
+) -> tuple[Categorical, npt.NDArray[np.double]]: ...

pandas-stubs/core/series.pyi

@@ -230,6 +230,16 @@ _ListLike: TypeAlias = (
 class Series(IndexOpsMixin[S1], NDFrame):
     __hash__: ClassVar[None]
 
+    @overload
+    def __new__(  # type: ignore[overload-overlap]
+        cls,
+        data: npt.NDArray[np.float64],
+        index: Axes | None = ...,
+        *,
+        dtype: Dtype = ...,
+        name: Hashable = ...,
+        copy: bool = ...,
+    ) -> Series[float]: ...
     @overload
     def __new__(  # type: ignore[overload-overlap]
         cls,

pandas-stubs/plotting/_core.pyi

@@ -340,7 +340,7 @@ class PlotAccessor:
             | Callable[[gaussian_kde], float]
             | None
         ) = ...,
-        ind: npt.NDArray[np.float_] | int | None = ...,
+        ind: npt.NDArray[np.double] | int | None = ...,
         *,
         subplots: Literal[False] | None = ...,
         **kwargs,
@@ -354,7 +354,7 @@ class PlotAccessor:
             | Callable[[gaussian_kde], float]
             | None
         ) = ...,
-        ind: npt.NDArray[np.float_] | int | None = ...,
+        ind: npt.NDArray[np.double] | int | None = ...,
         *,
         subplots: Literal[True],
         **kwargs,

pyproject.toml

@@ -8,22 +8,20 @@ readme = "README.md"
 homepage = "https://pandas.pydata.org"
 repository = "https://github.com/pandas-dev/pandas-stubs"
 classifiers = [
-    "Development Status :: 5 - Production/Stable",
-    "Environment :: Console",
-    "Intended Audience :: Science/Research",
-    "License :: OSI Approved :: BSD License",
-    "Operating System :: OS Independent",
-    "Programming Language :: Python",
-    "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3 :: Only",
-    "Programming Language :: Python :: 3.9",
-    "Programming Language :: Python :: 3.10",
-    "Programming Language :: Python :: 3.11",
-    "Topic :: Scientific/Engineering"
-]
-packages = [
-    { "include" = "pandas-stubs"}
+  "Development Status :: 5 - Production/Stable",
+  "Environment :: Console",
+  "Intended Audience :: Science/Research",
+  "License :: OSI Approved :: BSD License",
+  "Operating System :: OS Independent",
+  "Programming Language :: Python",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3 :: Only",
+  "Programming Language :: Python :: 3.9",
+  "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
+  "Topic :: Scientific/Engineering",
 ]
+packages = [{ "include" = "pandas-stubs" }]
 
 [tool.poetry.urls]
 "Bug Tracker" = "https://github.com/pandas-dev/pandas-stubs/issues"
@@ -33,26 +31,26 @@ packages = [
 python = ">=3.9"
 types-pytz = ">= 2022.1.1"
 numpy = [
-  { version = ">=1.23.5", python = ">=3.9,<3.12" },
-  { version = ">=1.26.0", python = ">=3.12,<3.13" }
+  { version = ">=1.23.5,<2.0.0", python = ">=3.9,<3.12" },
+  { version = ">=2.0.0", python = ">=3.12,<3.13" },
 ]
 
 [tool.poetry.group.dev.dependencies]
-mypy = "1.10.0"
+mypy = "1.10.1"
 pandas = "2.2.2"
 pyarrow = ">=10.0.1"
 pytest = ">=7.1.2"
-pyright = ">=1.1.365"
+pyright = ">=1.1.369"
 poethepoet = ">=0.16.5"
 loguru = ">=0.6.0"
 typing-extensions = ">=4.4.0"
-matplotlib = ">=3.5.1,<3.9.0"  # TODO https://github.com/pandas-dev/pandas/issues/58851
+matplotlib = ">=3.5.1,<3.9.0" # TODO https://github.com/pandas-dev/pandas/issues/58851
 pre-commit = ">=2.19.0"
 black = ">=23.3.0"
 isort = ">=5.12.0"
 openpyxl = ">=3.0.10"
 # for tables, MacOS gives random CI failures on 3.9.2
-tables = { version = "==3.9.2",  python = "<4"}  # 3.8.0 depends on blosc2 which caps python to <4
+tables = { version = "==3.9.2", python = "<4" }   # 3.8.0 depends on blosc2 which caps python to <4
 lxml = ">=4.9.1"
 pyreadstat = ">=1.2.0"
 xlrd = ">=2.0.1"
@@ -88,15 +86,19 @@ script = "scripts.test:test(dist=True)"
 [tool.poe.tasks.pytest]
 help = "Run pytest"
 script = "scripts.test:pytest(nightly)"
-args = [{name = "nightly", positional = false, default = false, type = "boolean", required = false, help= "Use pandas nightly (off by default)"}]
+args = [
+  { name = "nightly", positional = false, default = false, type = "boolean", required = false, help = "Use pandas nightly (off by default)" },
+]
 
 [tool.poe.tasks.style]
 help = "Run pre-commit"
 script = "scripts.test.run:style"
 
 [tool.poe.tasks.mypy]
 help = "Run mypy on 'tests' (using the local stubs) and on the local stubs"
-args = [{name = "mypy_nightly", positional = false, default = false, type = "boolean", required = false, help= "Use mypy nightly (off by default)"}]
+args = [
+  { name = "mypy_nightly", positional = false, default = false, type = "boolean", required = false, help = "Use mypy nightly (off by default)" },
+]
 script = "scripts.test:mypy_src(mypy_nightly)"
 
 [tool.poe.tasks.mypy_dist]
@@ -114,18 +116,38 @@ script = "scripts.test:test(dist=True, type_checker='pyright')"
 [tool.poe.tasks.stubtest]
 script = "scripts.test:stubtest(allowlist, check_missing, nightly)"
 help = "Run stubtest to compare the installed stubs against pandas"
-args = [{ name = "allowlist", positional = true, default = "", required = false, help= "Path to an allowlist (optional)" }, {name = "check_missing", positional = false, default = false, type = "boolean", required = false, help= "Report errors when the stubs are incomplete (off by default)"}, {name = "nightly", positional = false, default = false, type = "boolean", required = false, help= "Compare against pandas nightly (off by default)"}]
+args = [
+  { name = "allowlist", positional = true, default = "", required = false, help = "Path to an allowlist (optional)" },
+  { name = "check_missing", positional = false, default = false, type = "boolean", required = false, help = "Report errors when the stubs are incomplete (off by default)" },
+  { name = "nightly", positional = false, default = false, type = "boolean", required = false, help = "Compare against pandas nightly (off by default)" },
+]
 
 
 [tool.black]
 target-version = ['py39']
 
 [tool.isort]
 known_pre_libs = "pandas._config"
-known_pre_core = ["pandas._libs", "pandas._typing", "pandas.util._*", "pandas.compat", "pandas.errors"]
+known_pre_core = [
+  "pandas._libs",
+  "pandas._typing",
+  "pandas.util._*",
+  "pandas.compat",
+  "pandas.errors",
+]
 known_dtypes = "pandas.core.dtypes"
 known_post_core = ["pandas.tseries", "pandas.io", "pandas.plotting"]
-sections = ["FUTURE", "STDLIB", "THIRDPARTY" ,"PRE_LIBS" , "PRE_CORE", "DTYPES", "FIRSTPARTY", "POST_CORE", "LOCALFOLDER"]
+sections = [
+  "FUTURE",
+  "STDLIB",
+  "THIRDPARTY",
+  "PRE_LIBS",
+  "PRE_CORE",
+  "DTYPES",
+  "FIRSTPARTY",
+  "POST_CORE",
+  "LOCALFOLDER",
+]
 profile = "black"
 combine_as_imports = true
 force_grid_wrap = 2
@@ -142,16 +164,16 @@ follow_imports_for_stubs = false
 no_site_packages = false
 no_silence_site_packages = false
 # Disallow dynamic typing
-disallow_any_unimported = false # TODO
-disallow_any_expr = false # TODO
-disallow_any_decorated = false # TODO
-disallow_any_explicit = false # TODO
-disallow_any_generics = false # TODO
+disallow_any_unimported = false  # TODO
+disallow_any_expr = false        # TODO
+disallow_any_decorated = false   # TODO
+disallow_any_explicit = false    # TODO
+disallow_any_generics = false    # TODO
 disallow_subclassing_any = false # TODO
 # Untyped definitions and calls
-disallow_untyped_calls = false # TODO
-disallow_untyped_defs = false # TODO
-disallow_incomplete_defs = false # TODO
+disallow_untyped_calls = false     # TODO
+disallow_untyped_defs = false      # TODO
+disallow_incomplete_defs = false   # TODO
 check_untyped_defs = true
 disallow_untyped_decorators = true
 # None and Optional handling
@@ -161,16 +183,16 @@ strict_optional = true
 warn_redundant_casts = true
 warn_unused_ignores = true
 warn_no_return = true
-warn_return_any = false # TODO
-warn_unreachable = false # GH#27396
+warn_return_any = false     # TODO
+warn_unreachable = false    # GH#27396
 # Suppressing errors
 ignore_errors = false
 enable_error_code = "ignore-without-code" # same as in pandas
 # Miscellaneous strictness flags
 allow_untyped_globals = false
 allow_redefinition = false
 local_partial_types = false
-implicit_reexport = false # pyright behaves the same
+implicit_reexport = false     # pyright behaves the same
 strict_equality = true
 # Configuring error messages
 show_error_context = false
@@ -181,7 +203,7 @@ show_error_codes = true
 typeCheckingMode = "strict"
 stubPath = "."
 include = ["tests", "pandas-stubs"]
-enableTypeIgnoreComments = false  # use pyright-specific ignores
+enableTypeIgnoreComments = false    # use pyright-specific ignores
 # disable subset of strict
 reportMissingParameterType = false
 reportMissingTypeArgument = false

tests/__init__.py

@@ -13,6 +13,7 @@
     Literal,
 )
 
+import numpy as np
 import pandas as pd
 from pandas.core.groupby.groupby import BaseGroupBy
 from pandas.util.version import Version
@@ -23,6 +24,7 @@
 TYPE_CHECKING_INVALID_USAGE: Final = TYPE_CHECKING
 WINDOWS = os.name == "nt" or "cygwin" in platform.system().lower()
 PD_LTE_22 = Version(pd.__version__) < Version("2.2.999")
+NUMPY20 = np.lib.NumpyVersion(np.__version__) >= "2.0.0"
 
 
 def check(
@@ -40,7 +42,7 @@ def check(
     if isinstance(actual, pd.Series):
         value = actual.iloc[index_to_check_for_type]
     elif isinstance(actual, pd.Index):
-        value = actual[index_to_check_for_type]  # type: ignore[assignment]
+        value = actual[index_to_check_for_type]
     elif isinstance(actual, BaseGroupBy):
         value = actual.obj
     elif hasattr(actual, "__iter__"):

tests/test_io.py

@@ -53,6 +53,7 @@
     WINDOWS,
     check,
 )
+from tests import NUMPY20  # See https://github.com/PyTables/PyTables/issues/1172
 
 from pandas.io.api import to_pickle
 from pandas.io.json._json import JsonReader
@@ -347,12 +348,14 @@ def test_sas_xport() -> None:
         pass
 
 
+@pytest.mark.skipif(NUMPY20, reason="numpy 2.0 not compatible with Pytables")
 def test_hdf():
     with ensure_clean() as path:
         check(assert_type(DF.to_hdf(path, key="df"), None), type(None))
         check(assert_type(read_hdf(path), Union[DataFrame, Series]), DataFrame)
 
 
+@pytest.mark.skipif(NUMPY20, reason="numpy 2.0 not compatible with Pytables")
 def test_hdfstore() -> None:
     with ensure_clean() as path:
         store = HDFStore(path, model="w")
@@ -396,6 +399,7 @@ def test_hdfstore() -> None:
         store.close()
 
 
+@pytest.mark.skipif(NUMPY20, reason="numpy 2.0 not compatible with Pytables")
 def test_read_hdf_iterator() -> None:
     with ensure_clean() as path:
         check(assert_type(DF.to_hdf(path, key="df", format="table"), None), type(None))
@@ -410,6 +414,7 @@ def test_read_hdf_iterator() -> None:
         ti.close()
 
 
+@pytest.mark.skipif(NUMPY20, reason="numpy 2.0 not compatible with Pytables")
 def test_hdf_context_manager() -> None:
     with ensure_clean() as path:
         check(assert_type(DF.to_hdf(path, key="df", format="table"), None), type(None))
@@ -418,6 +423,7 @@ def test_hdf_context_manager() -> None:
             check(assert_type(store.get("df"), Union[DataFrame, Series]), DataFrame)
 
 
+@pytest.mark.skipif(NUMPY20, reason="numpy 2.0 not compatible with Pytables")
 def test_hdf_series():
     s = DF["a"]
     with ensure_clean() as path:

tests/test_pandas.py

@@ -1004,10 +1004,10 @@ def test_qcut() -> None:
     check(assert_type(c0, pd.Categorical), pd.Categorical)
     check(assert_type(d0, pd.Series), pd.Series)
 
-    check(assert_type(a1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(b1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(c1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(d1, npt.NDArray[np.float_]), np.ndarray)
+    check(assert_type(a1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(b1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(c1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(d1, npt.NDArray[np.double]), np.ndarray)
 
     e0, e1 = pd.qcut(val_list, [0.25, 0.5, 0.75], retbins=True)
     f0, f1 = pd.qcut(val_arr, np.array([0.25, 0.5, 0.75]), retbins=True)
@@ -1023,12 +1023,12 @@ def test_qcut() -> None:
     check(assert_type(i0, npt.NDArray[np.intp]), np.ndarray)
     check(assert_type(j0, npt.NDArray[np.intp]), np.ndarray)
 
-    check(assert_type(e1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(f1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(g1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(h1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(i1, npt.NDArray[np.float_]), np.ndarray)
-    check(assert_type(j1, npt.NDArray[np.float_]), np.ndarray)
+    check(assert_type(e1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(f1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(g1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(h1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(i1, npt.NDArray[np.double]), np.ndarray)
+    check(assert_type(j1, npt.NDArray[np.double]), np.ndarray)
 
 
 def test_merge() -> None: