TST/CLN: make equalContents more strict (pandas-dev#56154)

Author: mroeschke

pandas/_testing/__init__.py

@@ -302,13 +302,6 @@ def reset_display_options() -> None:
 # Comparators
 
 
-def equalContents(arr1, arr2) -> bool:
-    """
-    Checks if the set of unique elements of arr1 and arr2 are equivalent.
-    """
-    return frozenset(arr1) == frozenset(arr2)
-
-
 def box_expected(expected, box_cls, transpose: bool = True):
     """
     Helper function to wrap the expected output of a test in a given box_class.
@@ -1131,7 +1124,6 @@ def shares_memory(left, right) -> bool:
     "EMPTY_STRING_PATTERN",
     "ENDIAN",
     "ensure_clean",
-    "equalContents",
     "external_error_raised",
     "FLOAT_EA_DTYPES",
     "FLOAT_NUMPY_DTYPES",

pandas/tests/frame/indexing/test_indexing.py

@@ -48,7 +48,7 @@ def test_getitem(self, float_frame):
         # Column access
         for _, series in sl.items():
             assert len(series.index) == 20
-            assert tm.equalContents(series.index, sl.index)
+            tm.assert_index_equal(series.index, sl.index)
 
         for key, _ in float_frame._series.items():
             assert float_frame[key] is not None

pandas/tests/frame/methods/test_combine_first.py

@@ -37,7 +37,7 @@ def test_combine_first(self, float_frame):
         combined = head.combine_first(tail)
         reordered_frame = float_frame.reindex(combined.index)
         tm.assert_frame_equal(combined, reordered_frame)
-        assert tm.equalContents(combined.columns, float_frame.columns)
+        tm.assert_index_equal(combined.columns, float_frame.columns)
         tm.assert_series_equal(combined["A"], reordered_frame["A"])
 
         # same index

pandas/tests/frame/methods/test_reindex.py

@@ -624,7 +624,7 @@ def test_reindex(self, float_frame, using_copy_on_write):
                     assert np.isnan(val)
 
         for col, series in newFrame.items():
-            assert tm.equalContents(series.index, newFrame.index)
+            tm.assert_index_equal(series.index, newFrame.index)
         emptyFrame = float_frame.reindex(Index([]))
         assert len(emptyFrame.index) == 0
 
@@ -642,7 +642,7 @@ def test_reindex(self, float_frame, using_copy_on_write):
                     assert np.isnan(val)
 
         for col, series in nonContigFrame.items():
-            assert tm.equalContents(series.index, nonContigFrame.index)
+            tm.assert_index_equal(series.index, nonContigFrame.index)
 
         # corner cases
 

pandas/tests/frame/test_iteration.py

@@ -40,7 +40,7 @@ def test_items_names(self, float_string_frame):
             assert v.name == k
 
     def test_iter(self, float_frame):
-        assert tm.equalContents(list(float_frame), float_frame.columns)
+        assert list(float_frame) == list(float_frame.columns)
 
     def test_iterrows(self, float_frame, float_string_frame):
         for k, v in float_frame.iterrows():

pandas/tests/indexes/base_class/test_setops.py

@@ -12,6 +12,13 @@
 from pandas.core.algorithms import safe_sort
 
 
+def equal_contents(arr1, arr2) -> bool:
+    """
+    Checks if the set of unique elements of arr1 and arr2 are equivalent.
+    """
+    return frozenset(arr1) == frozenset(arr2)
+
+
 class TestIndexSetOps:
     @pytest.mark.parametrize(
         "method", ["union", "intersection", "difference", "symmetric_difference"]
@@ -71,7 +78,7 @@ def test_union_different_type_base(self, klass):
 
         result = first.union(klass(second.values))
 
-        assert tm.equalContents(result, index)
+        assert equal_contents(result, index)
 
     def test_union_sort_other_incomparable(self):
         # https://github.com/pandas-dev/pandas/issues/24959
@@ -119,7 +126,7 @@ def test_intersection_different_type_base(self, klass, sort):
         second = index[:3]
 
         result = first.intersection(klass(second.values), sort=sort)
-        assert tm.equalContents(result, second)
+        assert equal_contents(result, second)
 
     def test_intersection_nosort(self):
         result = Index(["c", "b", "a"]).intersection(["b", "a"])
@@ -244,7 +251,7 @@ def test_union_name_preservation(
             tm.assert_index_equal(union, expected)
         else:
             expected = Index(vals, name=expected_name)
-            tm.equalContents(union, expected)
+            tm.assert_index_equal(union.sort_values(), expected.sort_values())
 
     @pytest.mark.parametrize(
         "diff_type, expected",

pandas/tests/indexes/datetimes/test_setops.py

@@ -206,13 +206,13 @@ def test_intersection2(self):
         first = tm.makeDateIndex(10)
         second = first[5:]
         intersect = first.intersection(second)
-        assert tm.equalContents(intersect, second)
+        tm.assert_index_equal(intersect, second)
 
         # GH 10149
         cases = [klass(second.values) for klass in [np.array, Series, list]]
         for case in cases:
             result = first.intersection(case)
-            assert tm.equalContents(result, second)
+            tm.assert_index_equal(result, second)
 
         third = Index(["a", "b", "c"])
         result = first.intersection(third)

pandas/tests/indexes/interval/test_setops.py

@@ -25,14 +25,16 @@ def test_union(self, closed, sort):
 
         expected = monotonic_index(0, 13, closed=closed)
         result = index[::-1].union(other, sort=sort)
-        if sort is None:
+        if sort in (None, True):
             tm.assert_index_equal(result, expected)
-        assert tm.equalContents(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
 
         result = other[::-1].union(index, sort=sort)
-        if sort is None:
+        if sort in (None, True):
             tm.assert_index_equal(result, expected)
-        assert tm.equalContents(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
 
         tm.assert_index_equal(index.union(index, sort=sort), index)
         tm.assert_index_equal(index.union(index[:1], sort=sort), index)
@@ -65,14 +67,16 @@ def test_intersection(self, closed, sort):
 
         expected = monotonic_index(5, 11, closed=closed)
         result = index[::-1].intersection(other, sort=sort)
-        if sort is None:
+        if sort in (None, True):
             tm.assert_index_equal(result, expected)
-        assert tm.equalContents(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
 
         result = other[::-1].intersection(index, sort=sort)
-        if sort is None:
+        if sort in (None, True):
             tm.assert_index_equal(result, expected)
-        assert tm.equalContents(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
 
         tm.assert_index_equal(index.intersection(index, sort=sort), index)
 
@@ -148,16 +152,18 @@ def test_symmetric_difference(self, closed, sort):
         index = monotonic_index(0, 11, closed=closed)
         result = index[1:].symmetric_difference(index[:-1], sort=sort)
         expected = IntervalIndex([index[0], index[-1]])
-        if sort is None:
+        if sort in (None, True):
             tm.assert_index_equal(result, expected)
-        assert tm.equalContents(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
 
         # GH 19101: empty result, same dtype
         result = index.symmetric_difference(index, sort=sort)
         expected = empty_index(dtype="int64", closed=closed)
-        if sort is None:
+        if sort in (None, True):
             tm.assert_index_equal(result, expected)
-        assert tm.equalContents(result, expected)
+        else:
+            tm.assert_index_equal(result.sort_values(), expected)
 
         # GH 19101: empty result, different dtypes
         other = IntervalIndex.from_arrays(

pandas/tests/indexes/multi/test_setops.py

@@ -243,10 +243,10 @@ def test_union(idx, sort):
 
     the_union = piece1.union(piece2, sort=sort)
 
-    if sort is None:
-        tm.assert_index_equal(the_union, idx.sort_values())
-
-    assert tm.equalContents(the_union, idx)
+    if sort in (None, False):
+        tm.assert_index_equal(the_union.sort_values(), idx.sort_values())
+    else:
+        tm.assert_index_equal(the_union, idx)
 
     # corner case, pass self or empty thing:
     the_union = idx.union(idx, sort=sort)
@@ -258,7 +258,7 @@ def test_union(idx, sort):
     tuples = idx.values
     result = idx[:4].union(tuples[4:], sort=sort)
     if sort is None:
-        tm.equalContents(result, idx)
+        tm.assert_index_equal(result.sort_values(), idx.sort_values())
     else:
         assert result.equals(idx)
 
@@ -284,9 +284,10 @@ def test_intersection(idx, sort):
 
     the_int = piece1.intersection(piece2, sort=sort)
 
-    if sort is None:
+    if sort in (None, True):
         tm.assert_index_equal(the_int, idx[3:5])
-    assert tm.equalContents(the_int, idx[3:5])
+    else:
+        tm.assert_index_equal(the_int.sort_values(), idx[3:5])
 
     # corner case, pass self
     the_int = idx.intersection(idx, sort=sort)

pandas/tests/indexes/numeric/test_setops.py

@@ -133,7 +133,10 @@ def test_symmetric_difference(self, sort):
         index2 = Index([2, 3, 4, 1])
         result = index1.symmetric_difference(index2, sort=sort)
         expected = Index([5, 1])
-        assert tm.equalContents(result, expected)
+        if sort is not None:
+            tm.assert_index_equal(result, expected)
+        else:
+            tm.assert_index_equal(result, expected.sort_values())
         assert result.name is None
         if sort is None:
             expected = expected.sort_values()

pandas/tests/indexes/period/test_setops.py

@@ -142,9 +142,10 @@ def test_union_misc(self, sort):
 
         # not in order
         result = _permute(index[:-5]).union(_permute(index[10:]), sort=sort)
-        if sort is None:
+        if sort is False:
+            tm.assert_index_equal(result.sort_values(), index)
+        else:
             tm.assert_index_equal(result, index)
-        assert tm.equalContents(result, index)
 
         # cast if different frequencies
         index = period_range("1/1/2000", "1/20/2000", freq="D")
@@ -163,9 +164,10 @@ def test_intersection(self, sort):
         left = _permute(index[:-5])
         right = _permute(index[10:])
         result = left.intersection(right, sort=sort)
-        if sort is None:
+        if sort is False:
+            tm.assert_index_equal(result.sort_values(), index[10:-5])
+        else:
             tm.assert_index_equal(result, index[10:-5])
-        assert tm.equalContents(result, index[10:-5])
 
         # cast if different frequencies
         index = period_range("1/1/2000", "1/20/2000", freq="D")

pandas/tests/indexes/test_base.py

@@ -492,10 +492,10 @@ def test_union_dt_as_obj(self, simple_index):
         first_cat = index.union(date_index)
         second_cat = index.union(index)
 
-        appended = np.append(index, date_index.astype("O"))
+        appended = Index(np.append(index, date_index.astype("O")))
 
-        assert tm.equalContents(first_cat, appended)
-        assert tm.equalContents(second_cat, index)
+        tm.assert_index_equal(first_cat, appended)
+        tm.assert_index_equal(second_cat, index)
         tm.assert_contains_all(index, first_cat)
         tm.assert_contains_all(index, second_cat)
         tm.assert_contains_all(date_index, first_cat)