TST: Use more pytest fixtures (pandas-dev#53750)

* CI: Build pandas even if doctests fail

* BUG: groupby sum turning `inf+inf` and `(-inf)+(-inf)` into `nan` (pandas-dev#53623)

* DEPR: method, limit in NDFrame.replace (pandas-dev#53492)

* DEPR: method, limit in NDFrame.replace

* update test, docs

* suppress doctest warning

* doctests

* PERF: Series.str.get_dummies for ArrowDtype(pa.string()) (pandas-dev#53655)

* PERF: Series.str.get_dummies for ArrowDtype(pa.string())

* whatsnew

* typing

* TYP: core.missing (pandas-dev#53625)

* CI: Attempt to fix wheel builds (pandas-dev#53670)

* DOC: Fixing EX01 - Added examples (pandas-dev#53647)

* SeriesGroupBy.fillna example added

* Added examples

* Corrected failing test for timedelta.total_seconds

* Corrected fillna example

* CI/TST: Mark test_to_read_gcs as single_cpu (pandas-dev#53677)

* BUG/CoW: is_range_indexer can't handle very large arrays (pandas-dev#53672)

* BUG: is_range_indexer can't handle very large arrays

* fix test on 32-bit

* TST: Use more pytest fixtures

---------

Co-authored-by: Yao Xiao <[email protected]>
Co-authored-by: jbrockmendel <[email protected]>
Co-authored-by: Luke Manley <[email protected]>
Co-authored-by: Thomas Li <[email protected]>
Co-authored-by: Dea María Léon <[email protected]>

pandas/tests/resample/test_resample_api.py

@@ -16,38 +16,43 @@
 import pandas._testing as tm
 from pandas.core.indexes.datetimes import date_range
 
-dti = date_range(start=datetime(2005, 1, 1), end=datetime(2005, 1, 10), freq="Min")
 
-test_series = Series(np.random.rand(len(dti)), dti)
-_test_frame = DataFrame({"A": test_series, "B": test_series, "C": np.arange(len(dti))})
+@pytest.fixture
+def dti():
+    return date_range(start=datetime(2005, 1, 1), end=datetime(2005, 1, 10), freq="Min")
+
+
+@pytest.fixture
+def _test_series(dti):
+    return Series(np.random.rand(len(dti)), dti)
 
 
 @pytest.fixture
-def test_frame():
-    return _test_frame.copy()
+def test_frame(dti, _test_series):
+    return DataFrame({"A": _test_series, "B": _test_series, "C": np.arange(len(dti))})
 
 
-def test_str():
-    r = test_series.resample("H")
+def test_str(_test_series):
+    r = _test_series.resample("H")
     assert (
         "DatetimeIndexResampler [freq=<Hour>, axis=0, closed=left, "
         "label=left, convention=start, origin=start_day]" in str(r)
     )
 
-    r = test_series.resample("H", origin="2000-01-01")
+    r = _test_series.resample("H", origin="2000-01-01")
     assert (
         "DatetimeIndexResampler [freq=<Hour>, axis=0, closed=left, "
         "label=left, convention=start, origin=2000-01-01 00:00:00]" in str(r)
     )
 
 
-def test_api():
-    r = test_series.resample("H")
+def test_api(_test_series):
+    r = _test_series.resample("H")
     result = r.mean()
     assert isinstance(result, Series)
     assert len(result) == 217
 
-    r = test_series.to_frame().resample("H")
+    r = _test_series.to_frame().resample("H")
     result = r.mean()
     assert isinstance(result, DataFrame)
     assert len(result) == 217
@@ -116,11 +121,11 @@ def test_resample_group_keys():
     tm.assert_frame_equal(result, expected)
 
 
-def test_pipe(test_frame):
+def test_pipe(test_frame, _test_series):
     # GH17905
 
     # series
-    r = test_series.resample("H")
+    r = _test_series.resample("H")
     expected = r.max() - r.mean()
     result = r.pipe(lambda x: x.max() - x.mean())
     tm.assert_series_equal(result, expected)
@@ -261,9 +266,9 @@ def test_combined_up_downsampling_of_irregular():
     tm.assert_series_equal(result, expected)
 
 
-def test_transform_series():
-    r = test_series.resample("20min")
-    expected = test_series.groupby(pd.Grouper(freq="20min")).transform("mean")
+def test_transform_series(_test_series):
+    r = _test_series.resample("20min")
+    expected = _test_series.groupby(pd.Grouper(freq="20min")).transform("mean")
     result = r.transform("mean")
     tm.assert_series_equal(result, expected)
 
@@ -319,17 +324,17 @@ def test_fillna():
     ],
     ids=["resample", "groupby"],
 )
-def test_apply_without_aggregation(func):
+def test_apply_without_aggregation(func, _test_series):
     # both resample and groupby should work w/o aggregation
-    t = func(test_series)
+    t = func(_test_series)
     result = t.apply(lambda x: x)
-    tm.assert_series_equal(result, test_series)
+    tm.assert_series_equal(result, _test_series)
 
 
-def test_apply_without_aggregation2():
-    grouped = test_series.to_frame(name="foo").resample("20min", group_keys=False)
+def test_apply_without_aggregation2(_test_series):
+    grouped = _test_series.to_frame(name="foo").resample("20min", group_keys=False)
     result = grouped["foo"].apply(lambda x: x)
-    tm.assert_series_equal(result, test_series.rename("foo"))
+    tm.assert_series_equal(result, _test_series.rename("foo"))
 
 
 def test_agg_consistency():
@@ -1010,13 +1015,13 @@ def test_df_axis_param_depr():
         df.resample("M", axis=0)
 
 
-def test_series_axis_param_depr():
+def test_series_axis_param_depr(_test_series):
     warning_msg = (
         "The 'axis' keyword in Series.resample is "
         "deprecated and will be removed in a future version."
     )
     with tm.assert_produces_warning(FutureWarning, match=warning_msg):
-        test_series.resample("H", axis=0)
+        _test_series.resample("H", axis=0)
 
 
 def test_resample_empty():

pandas/tests/resample/test_resampler_grouper.py

@@ -17,10 +17,13 @@
 import pandas._testing as tm
 from pandas.core.indexes.datetimes import date_range
 
-test_frame = DataFrame(
-    {"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)},
-    index=date_range("1/1/2000", freq="s", periods=40),
-)
+
+@pytest.fixture
+def test_frame():
+    return DataFrame(
+        {"A": [1] * 20 + [2] * 12 + [3] * 8, "B": np.arange(40)},
+        index=date_range("1/1/2000", freq="s", periods=40),
+    )
 
 
 @async_mark()
@@ -85,7 +88,7 @@ def f_1(x):
     tm.assert_frame_equal(result, expected)
 
 
-def test_getitem():
+def test_getitem(test_frame):
     g = test_frame.groupby("A")
 
     expected = g.B.apply(lambda x: x.resample("2s").mean())
@@ -217,7 +220,7 @@ def test_nearest():
         "ohlc",
     ],
 )
-def test_methods(f):
+def test_methods(f, test_frame):
     g = test_frame.groupby("A")
     r = g.resample("2s")
 
@@ -226,7 +229,7 @@ def test_methods(f):
     tm.assert_equal(result, expected)
 
 
-def test_methods_nunique():
+def test_methods_nunique(test_frame):
     # series only
     g = test_frame.groupby("A")
     r = g.resample("2s")
@@ -236,15 +239,15 @@ def test_methods_nunique():
 
 
 @pytest.mark.parametrize("f", ["std", "var"])
-def test_methods_std_var(f):
+def test_methods_std_var(f, test_frame):
     g = test_frame.groupby("A")
     r = g.resample("2s")
     result = getattr(r, f)(ddof=1)
     expected = g.apply(lambda x: getattr(x.resample("2s"), f)(ddof=1))
     tm.assert_frame_equal(result, expected)
 
 
-def test_apply():
+def test_apply(test_frame):
     g = test_frame.groupby("A")
     r = g.resample("2s")
 
@@ -342,7 +345,7 @@ def test_resample_groupby_with_label():
     tm.assert_frame_equal(result, expected)
 
 
-def test_consistency_with_window():
+def test_consistency_with_window(test_frame):
     # consistent return values with window
     df = test_frame
     expected = Index([1, 2, 3], name="A")

pandas/tests/resample/test_time_grouper.py

@@ -14,10 +14,13 @@
 from pandas.core.groupby.grouper import Grouper
 from pandas.core.indexes.datetimes import date_range
 
-test_series = Series(np.random.randn(1000), index=date_range("1/1/2000", periods=1000))
 
+@pytest.fixture
+def test_series():
+    return Series(np.random.randn(1000), index=date_range("1/1/2000", periods=1000))
 
-def test_apply():
+
+def test_apply(test_series):
     grouper = Grouper(freq="A", label="right", closed="right")
 
     grouped = test_series.groupby(grouper)
@@ -33,7 +36,7 @@ def f(x):
     tm.assert_series_equal(applied, expected)
 
 
-def test_count():
+def test_count(test_series):
     test_series[::3] = np.nan
 
     expected = test_series.groupby(lambda x: x.year).count()
@@ -48,7 +51,7 @@ def test_count():
     tm.assert_series_equal(result, expected)
 
 
-def test_numpy_reduction():
+def test_numpy_reduction(test_series):
     result = test_series.resample("A", closed="right").prod()
 
     expected = test_series.groupby(lambda x: x.year).agg(np.prod)

pandas/tests/reshape/concat/test_append_common.py

@@ -10,51 +10,53 @@
 )
 import pandas._testing as tm
 
-dt_data = [
-    pd.Timestamp("2011-01-01"),
-    pd.Timestamp("2011-01-02"),
-    pd.Timestamp("2011-01-03"),
-]
-tz_data = [
-    pd.Timestamp("2011-01-01", tz="US/Eastern"),
-    pd.Timestamp("2011-01-02", tz="US/Eastern"),
-    pd.Timestamp("2011-01-03", tz="US/Eastern"),
-]
-td_data = [
-    pd.Timedelta("1 days"),
-    pd.Timedelta("2 days"),
-    pd.Timedelta("3 days"),
-]
-period_data = [
-    pd.Period("2011-01", freq="M"),
-    pd.Period("2011-02", freq="M"),
-    pd.Period("2011-03", freq="M"),
-]
-data_dict = {
-    "bool": [True, False, True],
-    "int64": [1, 2, 3],
-    "float64": [1.1, np.nan, 3.3],
-    "category": Categorical(["X", "Y", "Z"]),
-    "object": ["a", "b", "c"],
-    "datetime64[ns]": dt_data,
-    "datetime64[ns, US/Eastern]": tz_data,
-    "timedelta64[ns]": td_data,
-    "period[M]": period_data,
-}
+
+@pytest.fixture(
+    params=list(
+        {
+            "bool": [True, False, True],
+            "int64": [1, 2, 3],
+            "float64": [1.1, np.nan, 3.3],
+            "category": Categorical(["X", "Y", "Z"]),
+            "object": ["a", "b", "c"],
+            "datetime64[ns]": [
+                pd.Timestamp("2011-01-01"),
+                pd.Timestamp("2011-01-02"),
+                pd.Timestamp("2011-01-03"),
+            ],
+            "datetime64[ns, US/Eastern]": [
+                pd.Timestamp("2011-01-01", tz="US/Eastern"),
+                pd.Timestamp("2011-01-02", tz="US/Eastern"),
+                pd.Timestamp("2011-01-03", tz="US/Eastern"),
+            ],
+            "timedelta64[ns]": [
+                pd.Timedelta("1 days"),
+                pd.Timedelta("2 days"),
+                pd.Timedelta("3 days"),
+            ],
+            "period[M]": [
+                pd.Period("2011-01", freq="M"),
+                pd.Period("2011-02", freq="M"),
+                pd.Period("2011-03", freq="M"),
+            ],
+        }.items()
+    )
+)
+def item(request):
+    key, data = request.param
+    return key, data
+
+
+@pytest.fixture
+def item2(item):
+    return item
 
 
 class TestConcatAppendCommon:
     """
     Test common dtype coercion rules between concat and append.
     """
 
-    @pytest.fixture(params=sorted(data_dict.keys()))
-    def item(self, request):
-        key = request.param
-        return key, data_dict[key]
-
-    item2 = item
-
     def test_dtypes(self, item, index_or_series):
         # to confirm test case covers intended dtypes
         typ, vals = item

pandas/tests/series/methods/test_argsort.py

@@ -10,10 +10,11 @@
 
 
 class TestSeriesArgsort:
-    def _check_accum_op(self, name, ser, check_dtype=True):
-        func = getattr(np, name)
+    def test_argsort_numpy(self, datetime_series):
+        ser = datetime_series
+        func = np.argsort
         tm.assert_numpy_array_equal(
-            func(ser).values, func(np.array(ser)), check_dtype=check_dtype
+            func(ser).values, func(np.array(ser)), check_dtype=False
         )
 
         # with missing values
@@ -26,7 +27,6 @@ def _check_accum_op(self, name, ser, check_dtype=True):
         tm.assert_numpy_array_equal(result.values, expected, check_dtype=False)
 
     def test_argsort(self, datetime_series):
-        self._check_accum_op("argsort", datetime_series, check_dtype=False)
         argsorted = datetime_series.argsort()
         assert issubclass(argsorted.dtype.type, np.integer)