fix more tests

Author: jorisvandenbossche

pandas/core/groupby/grouper.py

@@ -12,7 +12,10 @@
 
 import numpy as np
 
-from pandas._config import using_copy_on_write
+from pandas._config import (
+    using_copy_on_write,
+    warn_copy_on_write,
+)
 
 from pandas._libs import lib
 from pandas._libs.tslibs import OutOfBoundsDatetime
@@ -966,7 +969,7 @@ def is_in_axis(key) -> bool:
     def is_in_obj(gpr) -> bool:
         if not hasattr(gpr, "name"):
             return False
-        if using_copy_on_write():
+        if using_copy_on_write or warn_copy_on_write():
             # For the CoW case, we check the references to determine if the
             # series is part of the object
             try:

pandas/tests/apply/test_frame_apply.py

@@ -1453,7 +1453,7 @@ def test_apply_dtype(col):
     tm.assert_series_equal(result, expected)
 
 
-def test_apply_mutating(using_array_manager, using_copy_on_write):
+def test_apply_mutating(using_array_manager, using_copy_on_write, warn_copy_on_write):
     # GH#35462 case where applied func pins a new BlockManager to a row
     df = DataFrame({"a": range(100), "b": range(100, 200)})
     df_orig = df.copy()
@@ -1467,7 +1467,8 @@ def func(row):
     expected = df.copy()
     expected["a"] += 1
 
-    result = df.apply(func, axis=1)
+    with tm.assert_cow_warning(warn_copy_on_write):
+        result = df.apply(func, axis=1)
 
     tm.assert_frame_equal(result, expected)
     if using_copy_on_write or using_array_manager:

pandas/tests/extension/base/setitem.py

@@ -405,7 +405,7 @@ def test_setitem_frame_2d_values(self, data):
         df.iloc[:] = df
         tm.assert_frame_equal(df, orig)
 
-        df.iloc[:-1] = df.iloc[:-1]
+        df.iloc[:-1] = df.iloc[:-1].copy()
         tm.assert_frame_equal(df, orig)
 
         df.iloc[:] = df.values

pandas/tests/generic/test_duplicate_labels.py

@@ -90,8 +90,10 @@ def test_preserve_getitem(self):
         assert df.loc[[0]].flags.allows_duplicate_labels is False
         assert df.loc[0, ["A"]].flags.allows_duplicate_labels is False
 
-    def test_ndframe_getitem_caching_issue(self, request, using_copy_on_write):
-        if not using_copy_on_write:
+    def test_ndframe_getitem_caching_issue(
+        self, request, using_copy_on_write, warn_copy_on_write
+    ):
+        if not (using_copy_on_write or warn_copy_on_write):
             request.applymarker(pytest.mark.xfail(reason="Unclear behavior."))
         # NDFrame.__getitem__ will cache the first df['A']. May need to
         # invalidate that cache? Update the cached entries?

pandas/tests/groupby/test_groupby.py

@@ -41,7 +41,9 @@ def test_repr():
     assert result == expected
 
 
-def test_groupby_std_datetimelike():
+# TODO(CoW-warn) this should NOT warn
+@pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
+def test_groupby_std_datetimelike(warn_copy_on_write):
     # GH#48481
     tdi = pd.timedelta_range("1 Day", periods=10000)
     ser = Series(tdi)

pandas/tests/indexes/period/test_partial_slicing.py

@@ -12,7 +12,9 @@
 
 
 class TestPeriodIndex:
-    def test_getitem_periodindex_duplicates_string_slice(self, using_copy_on_write):
+    def test_getitem_periodindex_duplicates_string_slice(
+        self, using_copy_on_write, warn_copy_on_write
+    ):
         # monotonic
         idx = PeriodIndex([2000, 2007, 2007, 2009, 2009], freq="Y-JUN")
         ts = Series(np.random.default_rng(2).standard_normal(len(idx)), index=idx)
@@ -21,7 +23,8 @@ def test_getitem_periodindex_duplicates_string_slice(self, using_copy_on_write):
         result = ts["2007"]
         expected = ts[1:3]
         tm.assert_series_equal(result, expected)
-        result[:] = 1
+        with tm.assert_cow_warning(warn_copy_on_write):
+            result[:] = 1
         if using_copy_on_write:
             tm.assert_series_equal(ts, original)
         else:

pandas/tests/indexing/multiindex/test_chaining_and_caching.py

@@ -42,7 +42,7 @@ def test_detect_chained_assignment(using_copy_on_write, warn_copy_on_write):
 
 
 @td.skip_array_manager_invalid_test  # with ArrayManager df.loc[0] is not a view
-def test_cache_updating(using_copy_on_write):
+def test_cache_updating(using_copy_on_write, warn_copy_on_write):
     # 5216
     # make sure that we don't try to set a dead cache
     a = np.random.default_rng(2).random((10, 3))
@@ -59,7 +59,8 @@ def test_cache_updating(using_copy_on_write):
             df.loc[0]["z"].iloc[0] = 1.0
         assert df.loc[(0, 0), "z"] == df_original.loc[0, "z"]
     else:
-        df.loc[0]["z"].iloc[0] = 1.0
+        with tm.assert_cow_warning(warn_copy_on_write):
+            df.loc[0]["z"].iloc[0] = 1.0
         result = df.loc[(0, 0), "z"]
         assert result == 1
 

pandas/tests/indexing/multiindex/test_partial.py

@@ -122,7 +122,10 @@ def test_getitem_partial_column_select(self):
     # exp.loc[2000, 4].values[:] select multiple columns -> .values is not a view
     @td.skip_array_manager_invalid_test
     def test_partial_set(
-        self, multiindex_year_month_day_dataframe_random_data, using_copy_on_write
+        self,
+        multiindex_year_month_day_dataframe_random_data,
+        using_copy_on_write,
+        warn_copy_on_write,
     ):
         # GH #397
         ymd = multiindex_year_month_day_dataframe_random_data
@@ -137,7 +140,8 @@ def test_partial_set(
                 df["A"].loc[2000, 4] = 1
             df.loc[(2000, 4), "A"] = 1
         else:
-            df["A"].loc[2000, 4] = 1
+            with tm.assert_cow_warning(warn_copy_on_write):
+                df["A"].loc[2000, 4] = 1
         exp.iloc[65:85, 0] = 1
         tm.assert_frame_equal(df, exp)
 
@@ -151,7 +155,8 @@ def test_partial_set(
                 df["A"].iloc[14] = 5
             df["A"].iloc[14] == exp["A"].iloc[14]
         else:
-            df["A"].iloc[14] = 5
+            with tm.assert_cow_warning(warn_copy_on_write):
+                df["A"].iloc[14] = 5
             assert df["A"].iloc[14] == 5
 
     @pytest.mark.parametrize("dtype", [int, float])

pandas/tests/indexing/multiindex/test_setitem.py

@@ -390,6 +390,7 @@ def test_loc_getitem_tuple_plus_columns(
         expected = df.loc[2000, 1, 6][["A", "B", "C"]]
         tm.assert_series_equal(result, expected)
 
+    @pytest.mark.filterwarnings("ignore:Setting a value on a view:FutureWarning")
     def test_loc_getitem_setitem_slice_integers(self, frame_or_series):
         index = MultiIndex(
             levels=[[0, 1, 2], [0, 2]], codes=[[0, 0, 1, 1, 2, 2], [0, 1, 0, 1, 0, 1]]
@@ -421,7 +422,7 @@ def test_setitem_change_dtype(self, multiindex_dataframe_random_data):
         tm.assert_series_equal(reindexed["foo", "two"], s > s.median())
 
     def test_set_column_scalar_with_loc(
-        self, multiindex_dataframe_random_data, using_copy_on_write
+        self, multiindex_dataframe_random_data, using_copy_on_write, warn_copy_on_write
     ):
         frame = multiindex_dataframe_random_data
         subset = frame.index[[1, 4, 5]]
@@ -431,7 +432,8 @@ def test_set_column_scalar_with_loc(
 
         frame_original = frame.copy()
         col = frame["B"]
-        col[subset] = 97
+        with tm.assert_cow_warning(warn_copy_on_write):
+            col[subset] = 97
         if using_copy_on_write:
             # chained setitem doesn't work with CoW
             tm.assert_frame_equal(frame, frame_original)

pandas/tests/indexing/multiindex/test_slice.py

@@ -739,6 +739,7 @@ def test_int_series_slicing(self, multiindex_year_month_day_dataframe_random_dat
         expected = s.reindex(s.index[5:])
         tm.assert_series_equal(result, expected)
 
+        s = ymd["A"].copy()
         exp = ymd["A"].copy()
         s[5:] = 0
         exp.iloc[5:] = 0

pandas/tests/indexing/test_chaining_and_caching.py

@@ -32,7 +32,9 @@ def random_text(nobs=100):
 
 
 class TestCaching:
-    def test_slice_consolidate_invalidate_item_cache(self, using_copy_on_write):
+    def test_slice_consolidate_invalidate_item_cache(
+        self, using_copy_on_write, warn_copy_on_write
+    ):
         # this is chained assignment, but will 'work'
         with option_context("chained_assignment", None):
             # #3970
@@ -49,7 +51,9 @@ def test_slice_consolidate_invalidate_item_cache(self, using_copy_on_write):
                 with tm.raises_chained_assignment_error():
                     df["bb"].iloc[0] = 0.17
             else:
-                df["bb"].iloc[0] = 0.17
+                # TODO(CoW-warn) custom warning message
+                with tm.assert_cow_warning(warn_copy_on_write):
+                    df["bb"].iloc[0] = 0.17
             df._clear_item_cache()
             if not using_copy_on_write:
                 tm.assert_almost_equal(df["bb"][0], 0.17)
@@ -74,7 +78,9 @@ def test_setitem_cache_updating(self, do_ref):
         assert df.loc[0, "c"] == 0.0
         assert df.loc[7, "c"] == 1.0
 
-    def test_setitem_cache_updating_slices(self, using_copy_on_write):
+    def test_setitem_cache_updating_slices(
+        self, using_copy_on_write, warn_copy_on_write
+    ):
         # GH 7084
         # not updating cache on series setting with slices
         expected = DataFrame(
@@ -102,7 +108,8 @@ def test_setitem_cache_updating_slices(self, using_copy_on_write):
                 with tm.raises_chained_assignment_error():
                     out[row["C"]][six:eix] = v
             else:
-                out[row["C"]][six:eix] = v
+                with tm.assert_cow_warning(warn_copy_on_write):
+                    out[row["C"]][six:eix] = v
 
         if not using_copy_on_write:
             tm.assert_frame_equal(out, expected)
@@ -113,17 +120,21 @@ def test_setitem_cache_updating_slices(self, using_copy_on_write):
 
         out = DataFrame({"A": [0, 0, 0]}, index=date_range("5/7/2014", "5/9/2014"))
         for ix, row in df.iterrows():
-            out.loc[six:eix, row["C"]] += row["D"]
+            # TODO(CoW-warn) should not warn
+            with tm.assert_produces_warning(FutureWarning):
+                out.loc[six:eix, row["C"]] += row["D"]
 
         tm.assert_frame_equal(out, expected)
         tm.assert_series_equal(out["A"], expected["A"])
 
-    def test_altering_series_clears_parent_cache(self, using_copy_on_write):
+    def test_altering_series_clears_parent_cache(
+        self, using_copy_on_write, warn_copy_on_write
+    ):
         # GH #33675
         df = DataFrame([[1, 2], [3, 4]], index=["a", "b"], columns=["A", "B"])
         ser = df["A"]
 
-        if using_copy_on_write:
+        if using_copy_on_write or warn_copy_on_write:
             assert "A" not in df._item_cache
         else:
             assert "A" in df._item_cache
@@ -138,7 +149,7 @@ def test_altering_series_clears_parent_cache(self, using_copy_on_write):
 
 
 class TestChaining:
-    def test_setitem_chained_setfault(self, using_copy_on_write):
+    def test_setitem_chained_setfault(self, using_copy_on_write, warn_copy_on_write):
         # GH6026
         data = ["right", "left", "left", "left", "right", "left", "timeout"]
         mdata = ["right", "left", "left", "left", "right", "left", "none"]
@@ -150,6 +161,8 @@ def test_setitem_chained_setfault(self, using_copy_on_write):
                 df.response[mask] = "none"
             tm.assert_frame_equal(df, DataFrame({"response": data}))
         else:
+            # TODO(CoW-warn) should warn
+            # with tm.assert_cow_warning(warn_copy_on_write):
             df.response[mask] = "none"
             tm.assert_frame_equal(df, DataFrame({"response": mdata}))
 
@@ -161,6 +174,8 @@ def test_setitem_chained_setfault(self, using_copy_on_write):
                 df.response[mask] = "none"
             tm.assert_frame_equal(df, DataFrame({"response": data}))
         else:
+            # TODO(CoW-warn) should warn
+            # with tm.assert_cow_warning(warn_copy_on_write):
             df.response[mask] = "none"
             tm.assert_frame_equal(df, DataFrame({"response": mdata}))
 
@@ -172,6 +187,8 @@ def test_setitem_chained_setfault(self, using_copy_on_write):
                 df.response[mask] = "none"
             tm.assert_frame_equal(df, df_original)
         else:
+            # TODO(CoW-warn) should warn
+            # with tm.assert_cow_warning(warn_copy_on_write):
             df.response[mask] = "none"
             tm.assert_frame_equal(df, DataFrame({"response": mdata, "response1": data}))
 
@@ -183,7 +200,8 @@ def test_setitem_chained_setfault(self, using_copy_on_write):
                 df["A"].iloc[0] = np.nan
             expected = DataFrame({"A": ["foo", "bar", "bah", "foo", "bar"]})
         else:
-            df["A"].iloc[0] = np.nan
+            with tm.assert_cow_warning(warn_copy_on_write):
+                df["A"].iloc[0] = np.nan
             expected = DataFrame({"A": [np.nan, "bar", "bah", "foo", "bar"]})
         result = df.head()
         tm.assert_frame_equal(result, expected)
@@ -193,7 +211,8 @@ def test_setitem_chained_setfault(self, using_copy_on_write):
             with tm.raises_chained_assignment_error():
                 df.A.iloc[0] = np.nan
         else:
-            df.A.iloc[0] = np.nan
+            with tm.assert_cow_warning(warn_copy_on_write):
+                df.A.iloc[0] = np.nan
         result = df.head()
         tm.assert_frame_equal(result, expected)
 
@@ -636,7 +655,9 @@ def test_cache_updating2(self, using_copy_on_write):
         expected = Series([0, 0, 0, 2, 0], name="f")
         tm.assert_series_equal(df.f, expected)
 
-    def test_iloc_setitem_chained_assignment(self, using_copy_on_write):
+    def test_iloc_setitem_chained_assignment(
+        self, using_copy_on_write, warn_copy_on_write
+    ):
         # GH#3970
         with option_context("chained_assignment", None):
             df = DataFrame({"aa": range(5), "bb": [2.2] * 5})
@@ -648,7 +669,8 @@ def test_iloc_setitem_chained_assignment(self, using_copy_on_write):
                 with tm.raises_chained_assignment_error():
                     df["bb"].iloc[0] = 0.13
             else:
-                df["bb"].iloc[0] = 0.13
+                with tm.assert_cow_warning(warn_copy_on_write):
+                    df["bb"].iloc[0] = 0.13
 
             # GH#3970 this lookup used to break the chained setting to 0.15
             df.iloc[ck]
@@ -657,7 +679,8 @@ def test_iloc_setitem_chained_assignment(self, using_copy_on_write):
                 with tm.raises_chained_assignment_error():
                     df["bb"].iloc[0] = 0.15
             else:
-                df["bb"].iloc[0] = 0.15
+                with tm.assert_cow_warning(warn_copy_on_write):
+                    df["bb"].iloc[0] = 0.15
 
             if not using_copy_on_write:
                 assert df["bb"].iloc[0] == 0.15

pandas/tests/indexing/test_iat.py

@@ -5,6 +5,7 @@
     Series,
     period_range,
 )
+import pandas._testing as tm
 
 
 def test_iat(float_frame):
@@ -30,17 +31,22 @@ def test_iat_getitem_series_with_period_index():
     assert expected == result
 
 
-def test_iat_setitem_item_cache_cleared(indexer_ial, using_copy_on_write):
+def test_iat_setitem_item_cache_cleared(
+    indexer_ial, using_copy_on_write, warn_copy_on_write
+):
     # GH#45684
     data = {"x": np.arange(8, dtype=np.int64), "y": np.int64(0)}
     df = DataFrame(data).copy()
     ser = df["y"]
 
     # previously this iat setting would split the block and fail to clear
     #  the item_cache.
-    indexer_ial(df)[7, 0] = 9999
+    with tm.assert_cow_warning(warn_copy_on_write and indexer_ial is tm.iloc):
+        indexer_ial(df)[7, 0] = 9999
 
-    indexer_ial(df)[7, 1] = 1234
+    # TODO(CoW-warn) should also warn for iat?
+    with tm.assert_cow_warning(warn_copy_on_write and indexer_ial is tm.iloc):
+        indexer_ial(df)[7, 1] = 1234
 
     assert df.iat[7, 1] == 1234
     if not using_copy_on_write: