fix additional tests that were relying on chained assignment

Author: jorisvandenbossche

pandas/tests/apply/test_frame_apply.py

@@ -347,7 +347,7 @@ def test_apply_yield_list(float_frame):
 
 
 def test_apply_reduce_Series(float_frame):
-    float_frame["A"].iloc[::2] = np.nan
+    float_frame.iloc[::2, float_frame.columns.get_loc("A")] = np.nan
     expected = float_frame.mean(1)
     result = float_frame.apply(np.mean, axis=1)
     tm.assert_series_equal(result, expected)

pandas/tests/frame/indexing/test_setitem.py

@@ -1136,7 +1136,7 @@ def test_setitem_always_copy(self, float_frame, using_copy_on_write):
         if using_copy_on_write:
             with pytest.raises(ChainedAssignmentError):
                 float_frame["E"][5:10] = np.nan
-            float_frame.loc[5:10, "E"] = np.nan
+            float_frame.loc[float_frame.index[5] : float_frame.index[9], "E"] = np.nan
         else:
             float_frame["E"][5:10] = np.nan
         assert notna(s[5:10]).all()

pandas/tests/frame/methods/test_cov_corr.py

@@ -107,9 +107,10 @@ class TestDataFrameCorr:
     @pytest.mark.parametrize("method", ["pearson", "kendall", "spearman"])
     @td.skip_if_no_scipy
     def test_corr_scipy_method(self, float_frame, method):
-        float_frame["A"][:5] = np.nan
-        float_frame["B"][5:10] = np.nan
-        float_frame["A"][:10] = float_frame["A"][10:20]
+        idx = float_frame.index
+        float_frame.loc[: idx[4], "A"] = np.nan
+        float_frame.loc[idx[5] : idx[9], "B"] = np.nan
+        float_frame.loc[: idx[9], "A"] = float_frame["A"][10:20]
 
         correls = float_frame.corr(method=method)
         expected = float_frame["A"].corr(float_frame["C"], method=method)

pandas/tests/frame/methods/test_fillna.py

@@ -392,16 +392,18 @@ def test_fillna_datetime_columns(self):
         tm.assert_frame_equal(result, expected)
 
     def test_ffill(self, datetime_frame):
-        datetime_frame["A"][:5] = np.nan
-        datetime_frame["A"][-5:] = np.nan
+        idx = datetime_frame.index
+        datetime_frame.loc[: idx[4], "A"] = np.nan
+        datetime_frame.loc[idx[-5] :, "A"] = np.nan
 
         tm.assert_frame_equal(
             datetime_frame.ffill(), datetime_frame.fillna(method="ffill")
         )
 
     def test_bfill(self, datetime_frame):
-        datetime_frame["A"][:5] = np.nan
-        datetime_frame["A"][-5:] = np.nan
+        idx = datetime_frame.index
+        datetime_frame.loc[: idx[4], "A"] = np.nan
+        datetime_frame.loc[idx[-5] :, "A"] = np.nan
 
         tm.assert_frame_equal(
             datetime_frame.bfill(), datetime_frame.fillna(method="bfill")
@@ -467,8 +469,8 @@ def test_fillna_integer_limit(self, type):
 
     def test_fillna_inplace(self):
         df = DataFrame(np.random.randn(10, 4))
-        df[1][:4] = np.nan
-        df[3][-4:] = np.nan
+        df.loc[:4, 1] = np.nan
+        df.loc[-4:, 3] = np.nan
 
         expected = df.fillna(value=0)
         assert expected is not df
@@ -479,8 +481,8 @@ def test_fillna_inplace(self):
         expected = df.fillna(value={0: 0}, inplace=True)
         assert expected is None
 
-        df[1][:4] = np.nan
-        df[3][-4:] = np.nan
+        df.loc[:4, 1] = np.nan
+        df.loc[-4:, 3] = np.nan
         expected = df.fillna(method="ffill")
         assert expected is not df
 

pandas/tests/frame/methods/test_rank.py

@@ -43,10 +43,10 @@ def test_rank(self, float_frame):
         import scipy.stats  # noqa:F401
         from scipy.stats import rankdata
 
-        float_frame["A"][::2] = np.nan
-        float_frame["B"][::3] = np.nan
-        float_frame["C"][::4] = np.nan
-        float_frame["D"][::5] = np.nan
+        float_frame.loc[::2, "A"] = np.nan
+        float_frame.loc[::3, "B"] = np.nan
+        float_frame.loc[::4, "C"] = np.nan
+        float_frame.loc[::5, "D"] = np.nan
 
         ranks0 = float_frame.rank()
         ranks1 = float_frame.rank(1)
@@ -148,10 +148,10 @@ def test_rank_na_option(self, float_frame):
         import scipy.stats  # noqa:F401
         from scipy.stats import rankdata
 
-        float_frame["A"][::2] = np.nan
-        float_frame["B"][::3] = np.nan
-        float_frame["C"][::4] = np.nan
-        float_frame["D"][::5] = np.nan
+        float_frame.loc[::2, "A"] = np.nan
+        float_frame.loc[::3, "B"] = np.nan
+        float_frame.loc[::4, "C"] = np.nan
+        float_frame.loc[::5, "D"] = np.nan
 
         # bottom
         ranks0 = float_frame.rank(na_option="bottom")

pandas/tests/frame/methods/test_replace.py

@@ -29,8 +29,9 @@ def mix_abc() -> dict[str, list[float | str]]:
 
 class TestDataFrameReplace:
     def test_replace_inplace(self, datetime_frame, float_string_frame):
-        datetime_frame["A"][:5] = np.nan
-        datetime_frame["A"][-5:] = np.nan
+        idx = datetime_frame.index
+        datetime_frame.loc[: idx[4], "A"] = np.nan
+        datetime_frame.loc[idx[-5] :, "A"] = np.nan
 
         tsframe = datetime_frame.copy()
         return_value = tsframe.replace(np.nan, 0, inplace=True)
@@ -420,16 +421,17 @@ def test_regex_replace_string_types(
         tm.assert_equal(result, expected)
 
     def test_replace(self, datetime_frame):
-        datetime_frame["A"][:5] = np.nan
-        datetime_frame["A"][-5:] = np.nan
+        idx = datetime_frame.index
+        datetime_frame.loc[: idx[4], "A"] = np.nan
+        datetime_frame.loc[idx[-5] :, "A"] = np.nan
 
         zero_filled = datetime_frame.replace(np.nan, -1e8)
         tm.assert_frame_equal(zero_filled, datetime_frame.fillna(-1e8))
         tm.assert_frame_equal(zero_filled.replace(-1e8, np.nan), datetime_frame)
 
-        datetime_frame["A"][:5] = np.nan
-        datetime_frame["A"][-5:] = np.nan
-        datetime_frame["B"][:5] = -1e8
+        datetime_frame.loc[: idx[4], "A"] = np.nan
+        datetime_frame.loc[idx[-5] :, "A"] = np.nan
+        datetime_frame.loc[: idx[4], "B"] = -1e8
 
         # empty
         df = DataFrame(index=["a", "b"])
@@ -716,16 +718,17 @@ def test_replace_for_new_dtypes(self, datetime_frame):
 
         # dtypes
         tsframe = datetime_frame.copy().astype(np.float32)
-        tsframe["A"][:5] = np.nan
-        tsframe["A"][-5:] = np.nan
+        idx = tsframe.index
+        tsframe.loc[: idx[4], "A"] = np.nan
+        tsframe.loc[idx[-5] :, "A"] = np.nan
 
         zero_filled = tsframe.replace(np.nan, -1e8)
         tm.assert_frame_equal(zero_filled, tsframe.fillna(-1e8))
         tm.assert_frame_equal(zero_filled.replace(-1e8, np.nan), tsframe)
 
-        tsframe["A"][:5] = np.nan
-        tsframe["A"][-5:] = np.nan
-        tsframe["B"][:5] = -1e8
+        tsframe.loc[: idx[4], "A"] = np.nan
+        tsframe.loc[idx[-5] :, "A"] = np.nan
+        tsframe.loc[: idx[4], "B"] = -1e8
 
         b = tsframe["B"]
         b[b == -1e8] = np.nan

pandas/tests/frame/methods/test_sort_values.py

@@ -330,13 +330,20 @@ def test_sort_values_datetimes(self):
         df2 = df.sort_values(by=["C", "B"])
         tm.assert_frame_equal(df1, df2)
 
-    def test_sort_values_frame_column_inplace_sort_exception(self, float_frame):
+    def test_sort_values_frame_column_inplace_sort_exception(
+        self, float_frame, using_copy_on_write
+    ):
         s = float_frame["A"]
-        with pytest.raises(ValueError, match="This Series is a view"):
+        if using_copy_on_write:
+            # with CoW, s is a new object detached from float_frame, so we can
+            # actually modify it inplace
             s.sort_values(inplace=True)
+        else:
+            with pytest.raises(ValueError, match="This Series is a view"):
+                s.sort_values(inplace=True)
 
-        cp = s.copy()
-        cp.sort_values()  # it works!
+            cp = s.copy()
+            cp.sort_values()  # it works!
 
     def test_sort_values_nat_values_in_int_column(self):
 

pandas/tests/frame/methods/test_to_csv.py

@@ -35,7 +35,7 @@ def read_csv(self, path, **kwargs):
     def test_to_csv_from_csv1(self, float_frame, datetime_frame):
 
         with tm.ensure_clean("__tmp_to_csv_from_csv1__") as path:
-            float_frame["A"][:5] = np.nan
+            float_frame.iloc[:5, float_frame.columns.get_loc("A")] = np.nan
 
             float_frame.to_csv(path)
             float_frame.to_csv(path, columns=["A", "B"])

pandas/tests/frame/test_block_internals.py

@@ -7,7 +7,10 @@
 import numpy as np
 import pytest
 
-from pandas.errors import PerformanceWarning
+from pandas.errors import (
+    ChainedAssignmentError,
+    PerformanceWarning,
+)
 import pandas.util._test_decorators as td
 
 import pandas as pd
@@ -346,7 +349,11 @@ def test_stale_cached_series_bug_473(self, using_copy_on_write):
             )
             repr(Y)
             Y["e"] = Y["e"].astype("object")
-            Y["g"]["c"] = np.NaN
+            if using_copy_on_write:
+                with pytest.raises(ChainedAssignmentError):
+                    Y["g"]["c"] = np.NaN
+            else:
+                Y["g"]["c"] = np.NaN
             repr(Y)
             result = Y.sum()  # noqa
             exp = Y["g"].sum()  # noqa

pandas/tests/frame/test_constructors.py

@@ -719,7 +719,8 @@ def test_constructor_defaultdict(self, float_frame):
         from collections import defaultdict
 
         data = {}
-        float_frame["B"][:10] = np.nan
+        float_frame.loc[: float_frame.index[10], "B"] = np.nan
+
         for k, v in float_frame.items():
             dct = defaultdict(dict)
             dct.update(v.to_dict())
@@ -2204,7 +2205,9 @@ def test_constructor_series_copy(self, float_frame):
         series = float_frame._series
 
         df = DataFrame({"A": series["A"]}, copy=True)
-        df["A"][:] = 5
+        # TODO can be replaced with `df.loc[:, "A"] = 5` after deprecation about
+        # inplace mutation is enforced
+        df.loc[df.index[0] : df.index[-1], "A"] = 5
 
         assert not (series["A"] == 5).all()
 

pandas/tests/io/excel/test_writers.py

@@ -374,7 +374,7 @@ def test_excel_writer_context_manager(self, frame, path):
 
     def test_roundtrip(self, frame, path):
         frame = frame.copy()
-        frame["A"][:5] = np.nan
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
 
         frame.to_excel(path, "test1")
         frame.to_excel(path, "test1", columns=["A", "B"])
@@ -444,7 +444,7 @@ def test_ts_frame(self, tsframe, path):
 
     def test_basics_with_nan(self, frame, path):
         frame = frame.copy()
-        frame["A"][:5] = np.nan
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
         frame.to_excel(path, "test1")
         frame.to_excel(path, "test1", columns=["A", "B"])
         frame.to_excel(path, "test1", header=False)
@@ -508,7 +508,7 @@ def test_sheets(self, frame, tsframe, path):
         tsframe.index = index
 
         frame = frame.copy()
-        frame["A"][:5] = np.nan
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
 
         frame.to_excel(path, "test1")
         frame.to_excel(path, "test1", columns=["A", "B"])
@@ -530,7 +530,7 @@ def test_sheets(self, frame, tsframe, path):
 
     def test_colaliases(self, frame, path):
         frame = frame.copy()
-        frame["A"][:5] = np.nan
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
 
         frame.to_excel(path, "test1")
         frame.to_excel(path, "test1", columns=["A", "B"])
@@ -548,7 +548,7 @@ def test_colaliases(self, frame, path):
 
     def test_roundtrip_indexlabels(self, merge_cells, frame, path):
         frame = frame.copy()
-        frame["A"][:5] = np.nan
+        frame.iloc[:5, frame.columns.get_loc("A")] = np.nan
 
         frame.to_excel(path, "test1")
         frame.to_excel(path, "test1", columns=["A", "B"])

pandas/tests/series/accessors/test_dt_accessor.py

@@ -12,7 +12,10 @@
 import pytz
 
 from pandas._libs.tslibs.timezones import maybe_get_tz
-from pandas.errors import SettingWithCopyError
+from pandas.errors import (
+    ChainedAssignmentError,
+    SettingWithCopyError,
+)
 
 from pandas.core.dtypes.common import (
     is_integer_dtype,
@@ -287,8 +290,8 @@ def test_dt_accessor_not_writeable(self, using_copy_on_write):
         msg = "modifications to a property of a datetimelike.+not supported"
         with pd.option_context("chained_assignment", "raise"):
             if using_copy_on_write:
-                # TODO(CoW) it would be nice to keep a warning/error for this case
-                ser.dt.hour[0] = 5
+                with pytest.raises(ChainedAssignmentError):
+                    ser.dt.hour[0] = 5
             else:
                 with pytest.raises(SettingWithCopyError, match=msg):
                     ser.dt.hour[0] = 5

pandas/tests/series/methods/test_rank.py

@@ -44,7 +44,7 @@ def test_rank(self, datetime_series):
         from scipy.stats import rankdata
 
         datetime_series[::2] = np.nan
-        datetime_series[:10][::3] = 4.0
+        datetime_series[:10:3] = 4.0
 
         ranks = datetime_series.rank()
         oranks = datetime_series.astype("O").rank()

pandas/tests/series/methods/test_sort_values.py

@@ -10,7 +10,7 @@
 
 
 class TestSeriesSortValues:
-    def test_sort_values(self, datetime_series):
+    def test_sort_values(self, datetime_series, using_copy_on_write):
 
         # check indexes are reordered corresponding with the values
         ser = Series([3, 2, 4, 1], ["A", "B", "C", "D"])
@@ -80,13 +80,18 @@ def test_sort_values(self, datetime_series):
         # Series.sort_values operating on a view
         df = DataFrame(np.random.randn(10, 4))
         s = df.iloc[:, 0]
+        s_orig_sorted = s.copy().sort_values()
 
-        msg = (
-            "This Series is a view of some other array, to sort in-place "
-            "you must create a copy"
-        )
-        with pytest.raises(ValueError, match=msg):
+        if using_copy_on_write:
             s.sort_values(inplace=True)
+            tm.assert_series_equal(s, s_orig_sorted)
+        else:
+            msg = (
+                "This Series is a view of some other array, to sort in-place "
+                "you must create a copy"
+            )
+            with pytest.raises(ValueError, match=msg):
+                s.sort_values(inplace=True)
 
     def test_sort_values_categorical(self):