Fix nightly CI again (#824)

* fix CI

* and removebytes from read_excel

* copy of getSeriesData

.pre-commit-config.yaml

@@ -7,11 +7,11 @@ repos:
     hooks:
     -   id: black
 -   repo: https://github.com/PyCQA/isort
-    rev: 5.12.0
+    rev: 5.13.0
     hooks:
     -   id: isort
 -   repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.1.6
+    rev: v0.1.7
     hooks:
     -   id: ruff
         args: [

pandas-stubs/_testing/__init__.pyi

@@ -12,7 +12,6 @@ from pandas import (
     Series,
 )
 
-def getSeriesData(): ...
 def assert_almost_equal(
     left,
     right,

pandas-stubs/api/types/__init__.pyi

@@ -5,37 +5,31 @@ from pandas.core.dtypes.api import (
     is_array_like as is_array_like,
     is_bool as is_bool,
     is_bool_dtype as is_bool_dtype,
-    is_categorical_dtype as is_categorical_dtype,
     is_complex as is_complex,
     is_complex_dtype as is_complex_dtype,
     is_datetime64_any_dtype as is_datetime64_any_dtype,
     is_datetime64_dtype as is_datetime64_dtype,
     is_datetime64_ns_dtype as is_datetime64_ns_dtype,
-    is_datetime64tz_dtype as is_datetime64tz_dtype,
     is_dict_like as is_dict_like,
     is_dtype_equal as is_dtype_equal,
     is_extension_array_dtype as is_extension_array_dtype,
     is_file_like as is_file_like,
     is_float as is_float,
     is_float_dtype as is_float_dtype,
     is_hashable as is_hashable,
-    is_int64_dtype as is_int64_dtype,
     is_integer as is_integer,
     is_integer_dtype as is_integer_dtype,
     is_interval as is_interval,
-    is_interval_dtype as is_interval_dtype,
     is_iterator as is_iterator,
     is_list_like as is_list_like,
     is_named_tuple as is_named_tuple,
     is_number as is_number,
     is_numeric_dtype as is_numeric_dtype,
     is_object_dtype as is_object_dtype,
-    is_period_dtype as is_period_dtype,
     is_re as is_re,
     is_re_compilable as is_re_compilable,
     is_scalar as is_scalar,
     is_signed_integer_dtype as is_signed_integer_dtype,
-    is_sparse as is_sparse,
     is_string_dtype as is_string_dtype,
     is_timedelta64_dtype as is_timedelta64_dtype,
     is_timedelta64_ns_dtype as is_timedelta64_ns_dtype,

pandas-stubs/core/dtypes/api.pyi

@@ -3,37 +3,31 @@ from pandas.core.dtypes.common import (
     is_array_like as is_array_like,
     is_bool as is_bool,
     is_bool_dtype as is_bool_dtype,
-    is_categorical_dtype as is_categorical_dtype,
     is_complex as is_complex,
     is_complex_dtype as is_complex_dtype,
     is_datetime64_any_dtype as is_datetime64_any_dtype,
     is_datetime64_dtype as is_datetime64_dtype,
     is_datetime64_ns_dtype as is_datetime64_ns_dtype,
-    is_datetime64tz_dtype as is_datetime64tz_dtype,
     is_dict_like as is_dict_like,
     is_dtype_equal as is_dtype_equal,
     is_extension_array_dtype as is_extension_array_dtype,
     is_file_like as is_file_like,
     is_float as is_float,
     is_float_dtype as is_float_dtype,
     is_hashable as is_hashable,
-    is_int64_dtype as is_int64_dtype,
     is_integer as is_integer,
     is_integer_dtype as is_integer_dtype,
     is_interval as is_interval,
-    is_interval_dtype as is_interval_dtype,
     is_iterator as is_iterator,
     is_list_like as is_list_like,
     is_named_tuple as is_named_tuple,
     is_number as is_number,
     is_numeric_dtype as is_numeric_dtype,
     is_object_dtype as is_object_dtype,
-    is_period_dtype as is_period_dtype,
     is_re as is_re,
     is_re_compilable as is_re_compilable,
     is_scalar as is_scalar,
     is_signed_integer_dtype as is_signed_integer_dtype,
-    is_sparse as is_sparse,
     is_string_dtype as is_string_dtype,
     is_timedelta64_dtype as is_timedelta64_dtype,
     is_timedelta64_ns_dtype as is_timedelta64_ns_dtype,

pandas-stubs/core/dtypes/common.pyi

@@ -32,19 +32,13 @@ _ArrayOrDtype: TypeAlias = (
 )
 
 def is_object_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
-def is_sparse(arr: ArrayLike | pd.Series | pd.DataFrame) -> bool: ...
 def is_datetime64_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
-def is_datetime64tz_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_timedelta64_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
-def is_period_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_dtype_equal(source, target) -> bool: ...
-def is_interval_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
-def is_categorical_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_string_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_integer_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_signed_integer_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_unsigned_integer_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
-def is_int64_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_datetime64_any_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_datetime64_ns_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...
 def is_timedelta64_ns_dtype(arr_or_dtype: _ArrayOrDtype) -> bool: ...

pandas-stubs/io/excel/_base.pyi

@@ -36,7 +36,6 @@ from pandas._typing import (
 def read_excel(
     io: FilePath
     | ReadBuffer[bytes]
-    | bytes
     | ExcelFile
     | Workbook
     | Book
@@ -75,7 +74,6 @@ def read_excel(
 def read_excel(
     io: FilePath
     | ReadBuffer[bytes]
-    | bytes
     | ExcelFile
     | Workbook
     | Book
@@ -115,7 +113,6 @@ def read_excel(
 def read_excel(  # type: ignore[misc]
     io: FilePath
     | ReadBuffer[bytes]
-    | bytes
     | ExcelFile
     | Workbook
     | Book
@@ -154,7 +151,6 @@ def read_excel(  # type: ignore[misc]
 def read_excel(
     io: FilePath
     | ReadBuffer[bytes]
-    | bytes
     | ExcelFile
     | Workbook
     | Book

tests/test_api_types.py

@@ -52,18 +52,6 @@ def test_is_bool_dtype() -> None:
     check(assert_type(api.is_bool_dtype(ExtensionDtype), bool), bool)
 
 
-def test_is_categorical_dtype() -> None:
-    with pytest_warns_bounded(
-        FutureWarning, "is_categorical_dtype is deprecated", lower="2.0.99"
-    ):
-        check(assert_type(api.is_categorical_dtype(arr), bool), bool)
-        check(assert_type(api.is_categorical_dtype(nparr), bool), bool)
-        check(assert_type(api.is_categorical_dtype(dtylike), bool), bool)
-        check(assert_type(api.is_categorical_dtype(dframe), bool), bool)
-        check(assert_type(api.is_categorical_dtype(ind), bool), bool)
-        check(assert_type(api.is_categorical_dtype(ExtensionDtype), bool), bool)
-
-
 def test_is_complex() -> None:
     check(assert_type(api.is_complex(obj), bool), bool)
     check(assert_type(api.is_complex(nparr), bool), bool)
@@ -124,18 +112,6 @@ def test_is_datetime64_ns_dtype() -> None:
     check(assert_type(api.is_datetime64_ns_dtype(ExtensionDtype), bool), bool)
 
 
-def test_is_datetime64tz_dtype() -> None:
-    with pytest_warns_bounded(
-        FutureWarning, "is_datetime64tz_dtype is deprecated", lower="2.0.99"
-    ):
-        check(assert_type(api.is_datetime64tz_dtype(arr), bool), bool)
-        check(assert_type(api.is_datetime64tz_dtype(nparr), bool), bool)
-        check(assert_type(api.is_datetime64tz_dtype(dtylike), bool), bool)
-        check(assert_type(api.is_datetime64tz_dtype(dframe), bool), bool)
-        check(assert_type(api.is_datetime64tz_dtype(ind), bool), bool)
-        check(assert_type(api.is_datetime64tz_dtype(ExtensionDtype), bool), bool)
-
-
 def test_is_dict_like() -> None:
     check(assert_type(api.is_dict_like(mapping), bool), bool)
     check(assert_type(api.is_dict_like(nparr), bool), bool)
@@ -209,18 +185,6 @@ def test_is_hashable() -> None:
     check(assert_type(api.is_hashable(ind), bool), bool)
 
 
-def test_is_int64_dtype() -> None:
-    with pytest_warns_bounded(
-        FutureWarning, "is_int64_dtype is deprecated", lower="2.0.99"
-    ):
-        check(assert_type(api.is_int64_dtype(arr), bool), bool)
-        check(assert_type(api.is_int64_dtype(nparr), bool), bool)
-        check(assert_type(api.is_int64_dtype(dtylike), bool), bool)
-        check(assert_type(api.is_int64_dtype(dframe), bool), bool)
-        check(assert_type(api.is_int64_dtype(ind), bool), bool)
-        # check(assert_type(api.is_int64_dtype(ExtensionDtype), bool), bool) pandas GH 50923
-
-
 def test_is_integer() -> None:
     check(assert_type(api.is_integer(obj), bool), bool)
     check(assert_type(api.is_integer(nparr), bool), bool)
@@ -260,19 +224,6 @@ def test_is_interval() -> None:
         check(assert_type(api.is_interval(ind), bool), bool)
 
 
-def test_is_interval_dtype() -> None:
-    with pytest_warns_bounded(
-        FutureWarning, "is_interval_dtype is deprecated", lower="2.0.99"
-    ):
-        check(assert_type(api.is_interval_dtype(obj), bool), bool)
-        check(assert_type(api.is_interval_dtype(nparr), bool), bool)
-        check(assert_type(api.is_interval_dtype(dtylike), bool), bool)
-        check(assert_type(api.is_interval_dtype(arr), bool), bool)
-        check(assert_type(api.is_interval_dtype(dframe), bool), bool)
-        check(assert_type(api.is_interval_dtype(ind), bool), bool)
-        check(assert_type(api.is_interval_dtype(ExtensionDtype), bool), bool)
-
-
 def test_is_iterator() -> None:
     check(assert_type(api.is_iterator(obj), bool), bool)
     check(assert_type(api.is_iterator(nparr), bool), bool)
@@ -342,18 +293,6 @@ def test_is_object_dtype() -> None:
     # check(assert_type(api.is_object_dtype(ExtensionDtype), bool), bool) pandas GH 50923
 
 
-def test_is_period_dtype() -> None:
-    with pytest_warns_bounded(
-        FutureWarning, "is_period_dtype is deprecated", lower="2.0.99"
-    ):
-        check(assert_type(api.is_period_dtype(arr), bool), bool)
-        check(assert_type(api.is_period_dtype(nparr), bool), bool)
-        check(assert_type(api.is_period_dtype(dtylike), bool), bool)
-        check(assert_type(api.is_period_dtype(dframe), bool), bool)
-        check(assert_type(api.is_period_dtype(ind), bool), bool)
-        check(assert_type(api.is_period_dtype(ExtensionDtype), bool), bool)
-
-
 def test_is_re() -> None:
     check(assert_type(api.is_re(obj), bool), bool)
     check(assert_type(api.is_re(nparr), bool), bool)
@@ -396,13 +335,6 @@ def test_is_signed_integer_dtype() -> None:
     # check(assert_type(api.is_signed_integer_dtype(ExtensionDtype), bool), bool) pandas GH 50923
 
 
-def test_is_sparse() -> None:
-    with pytest_warns_bounded(FutureWarning, "is_sparse is deprecated", lower="2.0.99"):
-        check(assert_type(api.is_sparse(arr), bool), bool)
-        check(assert_type(api.is_sparse(nparr), bool), bool)
-        check(assert_type(api.is_sparse(dframe), bool), bool)
-
-
 def test_is_string_dtype() -> None:
     check(assert_type(api.is_string_dtype(arr), bool), bool)
     check(assert_type(api.is_string_dtype(nparr), bool), bool)

tests/test_frame.py

@@ -14,6 +14,7 @@
 import io
 import itertools
 from pathlib import Path
+import string
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -28,10 +29,7 @@
 import numpy as np
 import numpy.typing as npt
 import pandas as pd
-from pandas._testing import (
-    ensure_clean,
-    getSeriesData,
-)
+from pandas._testing import ensure_clean
 from pandas.core.resample import Resampler  # noqa: F401
 from pandas.core.series import Series
 import pytest
@@ -52,6 +50,36 @@
 DF = pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]})
 
 
+def getCols(k) -> str:
+    return string.ascii_uppercase[:k]
+
+
+def makeStringIndex(k: int = 10) -> pd.Index:
+    return pd.Index(rands_array(nchars=10, size=k), name=None)
+
+
+def rands_array(nchars, size: int) -> np.ndarray:
+    chars = np.array(list(string.ascii_letters + string.digits), dtype=(np.str_, 1))
+    retval = (
+        np.random.default_rng(2)
+        .choice(chars, size=nchars * np.prod(size), replace=True)
+        .view((np.str_, nchars))
+        .reshape(size)
+    )
+    return retval.astype("O")
+
+
+def getSeriesData() -> dict[str, pd.Series]:
+    _N = 30
+    _K = 4
+
+    index = makeStringIndex(_N)
+    return {
+        c: pd.Series(np.random.default_rng(i).standard_normal(_N), index=index)
+        for i, c in enumerate(getCols(_K))
+    }
+
+
 def test_types_init() -> None:
     pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]})
     pd.DataFrame(data={"col1": [1, 2], "col2": [3, 4]}, index=[2, 1])

tests/test_io.py

@@ -469,8 +469,17 @@ def test_hdf_series():
 
 def test_spss():
     path = Path(CWD, "data", "labelled-num.sav")
-    check(assert_type(read_spss(path, convert_categoricals=True), DataFrame), DataFrame)
-    check(assert_type(read_spss(str(path), usecols=["VAR00002"]), DataFrame), DataFrame)
+    with pytest_warns_bounded(
+        FutureWarning, "ChainedAssignmentError: behaviour will change", lower="2.1.99"
+    ):
+        check(
+            assert_type(read_spss(path, convert_categoricals=True), DataFrame),
+            DataFrame,
+        )
+        check(
+            assert_type(read_spss(str(path), usecols=["VAR00002"]), DataFrame),
+            DataFrame,
+        )
 
 
 def test_json():
@@ -1022,14 +1031,6 @@ def test_read_excel_io_types() -> None:
         with as_path.open("rb") as as_file:
             check(assert_type(pd.read_excel(as_file), pd.DataFrame), pd.DataFrame)
 
-        as_bytes = as_path.read_bytes()
-        with pytest_warns_bounded(
-            FutureWarning,
-            "Passing bytes to 'read_excel' is deprecated",
-            lower="2.0.99",
-        ):
-            check(assert_type(pd.read_excel(as_bytes), pd.DataFrame), pd.DataFrame)
-
 
 def test_read_excel_basic():
     with ensure_clean(".xlsx") as path:

tests/test_scalars.py

@@ -844,7 +844,10 @@ def test_timedelta_cmp() -> None:
     c_dt_timedelta = dt.timedelta(days=1)
     c_timedelta64 = np.timedelta64(1, "D")
     c_ndarray_td64 = ndarray_td64
-    c_timedelta_index = pd.TimedeltaIndex([1, 2, 3], unit="D")
+    with pytest_warns_bounded(
+        FutureWarning, "The 'unit' keyword in TimedeltaIndex", lower="2.1.99"
+    ):
+        c_timedelta_index = pd.TimedeltaIndex([1, 2, 3], unit="D")
     c_timedelta_series = pd.Series(pd.TimedeltaIndex([1, 2, 3]))
 
     check(assert_type(td < c_timedelta, bool), bool)
@@ -1047,7 +1050,10 @@ def test_timedelta_cmp_rhs() -> None:
     c_dt_timedelta = dt.timedelta(days=1)
     c_timedelta64 = np.timedelta64(1, "D")
     c_ndarray_td64 = ndarray_td64
-    c_timedelta_index = pd.TimedeltaIndex([1, 2, 3], unit="D")
+    with pytest_warns_bounded(
+        FutureWarning, "The 'unit' keyword in TimedeltaIndex", lower="2.1.99"
+    ):
+        c_timedelta_index = pd.TimedeltaIndex([1, 2, 3], unit="D")
     c_timedelta_series = pd.Series(pd.TimedeltaIndex([1, 2, 3]))
 
     eq = check(assert_type(c_dt_timedelta == td, bool), bool)
@@ -1192,7 +1198,11 @@ def test_timestamp_add_sub() -> None:
     as_pd_timedelta = pd.Timedelta(days=1)
     as_dt_timedelta = dt.timedelta(days=1)
     as_offset = 3 * Day()
-    as_timedelta_index = pd.TimedeltaIndex([1, 2, 3], "D")
+
+    with pytest_warns_bounded(
+        FutureWarning, "The 'unit' keyword in TimedeltaIndex", lower="2.1.99"
+    ):
+        as_timedelta_index = pd.TimedeltaIndex([1, 2, 3], "D")
     as_timedelta_series = pd.Series(as_timedelta_index)
     check(assert_type(as_timedelta_series, TimedeltaSeries), pd.Series, pd.Timedelta)
     as_np_ndarray_td64 = np_td64_arr