TST: Split / parameterize resample pt2 (#45282)

Author: mroeschke

pandas/tests/resample/test_resample_api.py

@@ -142,21 +142,21 @@ def test_attribute_access(test_frame):
     tm.assert_series_equal(r.A.sum(), r["A"].sum())
 
 
-def test_api_compat_before_use():
+@pytest.mark.parametrize("attr", ["groups", "ngroups", "indices"])
+def test_api_compat_before_use(attr):
 
     # make sure that we are setting the binner
     # on these attributes
-    for attr in ["groups", "ngroups", "indices"]:
-        rng = date_range("1/1/2012", periods=100, freq="S")
-        ts = Series(np.arange(len(rng)), index=rng)
-        rs = ts.resample("30s")
+    rng = date_range("1/1/2012", periods=100, freq="S")
+    ts = Series(np.arange(len(rng)), index=rng)
+    rs = ts.resample("30s")
 
-        # before use
-        getattr(rs, attr)
+    # before use
+    getattr(rs, attr)
 
-        # after grouper is initialized is ok
-        rs.mean()
-        getattr(rs, attr)
+    # after grouper is initialized is ok
+    rs.mean()
+    getattr(rs, attr)
 
 
 def tests_skip_nuisance(test_frame):
@@ -273,15 +273,16 @@ def test_fillna():
         r.fillna(0)
 
 
-def test_apply_without_aggregation():
-
+@pytest.mark.parametrize(
+    "func",
+    [lambda x: x.resample("20min"), lambda x: x.groupby(pd.Grouper(freq="20min"))],
+    ids=["resample", "groupby"],
+)
+def test_apply_without_aggregation(func):
     # both resample and groupby should work w/o aggregation
-    r = test_series.resample("20min")
-    g = test_series.groupby(pd.Grouper(freq="20min"))
-
-    for t in [g, r]:
-        result = t.apply(lambda x: x)
-        tm.assert_series_equal(result, test_series)
+    t = func(test_series)
+    result = t.apply(lambda x: x)
+    tm.assert_series_equal(result, test_series)
 
 
 def test_agg_consistency():

pandas/tests/resample/test_resampler_grouper.py

@@ -202,44 +202,51 @@ def test_nearest():
     tm.assert_series_equal(result, expected)
 
 
-def test_methods():
+@pytest.mark.parametrize(
+    "f",
+    [
+        "first",
+        "last",
+        "median",
+        "sem",
+        "sum",
+        "mean",
+        "min",
+        "max",
+        "size",
+        "count",
+        "nearest",
+        "bfill",
+        "ffill",
+        "asfreq",
+        "ohlc",
+    ],
+)
+def test_methods(f):
     g = test_frame.groupby("A")
     r = g.resample("2s")
 
-    for f in ["first", "last", "median", "sem", "sum", "mean", "min", "max"]:
-        result = getattr(r, f)()
-        expected = g.apply(lambda x: getattr(x.resample("2s"), f)())
-        tm.assert_frame_equal(result, expected)
-
-    for f in ["size"]:
-        result = getattr(r, f)()
-        expected = g.apply(lambda x: getattr(x.resample("2s"), f)())
-        tm.assert_series_equal(result, expected)
+    result = getattr(r, f)()
+    expected = g.apply(lambda x: getattr(x.resample("2s"), f)())
+    tm.assert_equal(result, expected)
 
-    for f in ["count"]:
-        result = getattr(r, f)()
-        expected = g.apply(lambda x: getattr(x.resample("2s"), f)())
-        tm.assert_frame_equal(result, expected)
 
+def test_methods_nunique():
     # series only
-    for f in ["nunique"]:
-        result = getattr(r.B, f)()
-        expected = g.B.apply(lambda x: getattr(x.resample("2s"), f)())
-        tm.assert_series_equal(result, expected)
-
-    for f in ["nearest", "bfill", "ffill", "asfreq"]:
-        result = getattr(r, f)()
-        expected = g.apply(lambda x: getattr(x.resample("2s"), f)())
-        tm.assert_frame_equal(result, expected)
-
-    result = r.ohlc()
-    expected = g.apply(lambda x: x.resample("2s").ohlc())
-    tm.assert_frame_equal(result, expected)
+    g = test_frame.groupby("A")
+    r = g.resample("2s")
+    result = r.B.nunique()
+    expected = g.B.apply(lambda x: x.resample("2s").nunique())
+    tm.assert_series_equal(result, expected)
+
 
-    for f in ["std", "var"]:
-        result = getattr(r, f)(ddof=1)
-        expected = g.apply(lambda x: getattr(x.resample("2s"), f)(ddof=1))
-        tm.assert_frame_equal(result, expected)
+@pytest.mark.parametrize("f", ["std", "var"])
+def test_methods_std_var(f):
+    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():

pandas/tests/resample/test_time_grouper.py

@@ -145,16 +145,31 @@ def test_aggregate_normal(resample_method):
     expected.index = date_range(start="2013-01-01", freq="D", periods=5, name="key")
     tm.assert_equal(expected, dt_result)
 
-    # if TimeGrouper is used included, 'nth' doesn't work yet
-
-    """
-    for func in ['nth']:
-        expected = getattr(normal_grouped, func)(3)
-        expected.index = date_range(start='2013-01-01',
-                                    freq='D', periods=5, name='key')
-        dt_result = getattr(dt_grouped, func)(3)
-        tm.assert_frame_equal(expected, dt_result)
-    """
+
+@pytest.mark.xfail(reason="if TimeGrouper is used included, 'nth' doesn't work yet")
+def test_aggregate_nth(resample_method):
+    """Check TimeGrouper's aggregation is identical as normal groupby."""
+
+    data = np.random.randn(20, 4)
+    normal_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    normal_df["key"] = [1, 2, 3, 4, 5] * 4
+
+    dt_df = DataFrame(data, columns=["A", "B", "C", "D"])
+    dt_df["key"] = [
+        datetime(2013, 1, 1),
+        datetime(2013, 1, 2),
+        datetime(2013, 1, 3),
+        datetime(2013, 1, 4),
+        datetime(2013, 1, 5),
+    ] * 4
+
+    normal_grouped = normal_df.groupby("key")
+    dt_grouped = dt_df.groupby(Grouper(key="key", freq="D"))
+
+    expected = normal_grouped.nth(3)
+    expected.index = date_range(start="2013-01-01", freq="D", periods=5, name="key")
+    dt_result = dt_grouped.nth(3)
+    tm.assert_frame_equal(expected, dt_result)
 
 
 @pytest.mark.parametrize(