Adjust where the nulls should be, generalize some tests

Author: mroeschke

pandas/tests/arrays/boolean/test_arithmetic.py

@@ -11,7 +11,7 @@
 def data():
     """Fixture returning boolean array with valid and missing values."""
     return pd.array(
-        [True, False] + [np.nan] + [True, False] * 2 + [np.nan] + [True, False],
+        [True, False] * 2 + [np.nan] + [True, False] + [np.nan] + [True, False],
         dtype="boolean",
     )
 

pandas/tests/arrays/boolean/test_comparison.py

@@ -11,7 +11,7 @@
 def data():
     """Fixture returning boolean array with valid and missing data"""
     return pd.array(
-        [True, False] + [np.nan] + [True, False] * 2 + [np.nan] + [True, False],
+        [True, False] * 2 + [np.nan] + [True, False] + [np.nan] + [True, False],
         dtype="boolean",
     )
 

pandas/tests/arrays/boolean/test_reduction.py

@@ -8,7 +8,7 @@
 def data():
     """Fixture returning boolean array, with valid and missing values."""
     return pd.array(
-        [True, False] + [np.nan] + [True, False] * 2 + [np.nan] + [True, False],
+        [True, False] * 2 + [np.nan] + [True, False] + [np.nan] + [True, False],
         dtype="boolean",
     )
 

pandas/tests/arrays/floating/conftest.py

@@ -17,7 +17,7 @@ def dtype(request):
 def data(dtype):
     """Fixture returning 'data' array according to parametrized float 'dtype'"""
     return pd.array(
-        [0.1, 0.2] + [pd.NA] + [1.0, 1.1, 1.2, 1.3] + [pd.NA] + [9.9, 10.0],
+        [0.1, 0.2, 0.3, 0.4] + [pd.NA] + [1.0, 1.1] + [pd.NA] + [9.9, 10.0],
         dtype=dtype,
     )
 

pandas/tests/arrays/integer/conftest.py

@@ -39,7 +39,7 @@ def data(dtype):
     Used to test dtype conversion with and without missing values.
     """
     return pd.array(
-        [0, 1] + [pd.NA] + [10, 11, 12, 13] + [pd.NA] + [99, 100],
+        [0, 1, 2, 3] + [pd.NA] + [10, 11] + [pd.NA] + [99, 100],
         dtype=dtype,
     )
 

pandas/tests/extension/base/casting.py

@@ -57,9 +57,17 @@ def test_astype_str(self, data):
     )
     def test_astype_string(self, data, nullable_string_dtype):
         # GH-33465, GH#45326 as of 2.0 we decode bytes instead of calling str(obj)
+        def as_str(x):
+            if isinstance(x, bytes):
+                return x.decode()
+            elif x is data.dtype.na_value:
+                return x
+            else:
+                return str(x)
+
         result = pd.Series(data[:5]).astype(nullable_string_dtype)
         expected = pd.Series(
-            [str(x) if not isinstance(x, bytes) else x.decode() for x in data[:5]],
+            [as_str(x) for x in data[:5]],
             dtype=nullable_string_dtype,
         )
         tm.assert_series_equal(result, expected)

pandas/tests/extension/base/printing.py

@@ -13,7 +13,7 @@ def test_array_repr(self, data, size):
         if size == "small":
             data = data[:5]
         else:
-            data = type(data)._concat_same_type([data] * 5)
+            data = type(data)._concat_same_type([data] * 20)
 
         result = repr(data)
         assert type(data).__name__ in result

pandas/tests/extension/base/reshaping.py

@@ -303,7 +303,9 @@ def test_stack(self, data, columns, future_stack):
     )
     @pytest.mark.parametrize("obj", ["series", "frame"])
     def test_unstack(self, data, index, obj):
-        data = data[: len(index)]
+        final_length = min(len(index), len(data))
+        index = index[:final_length]
+        data = data[:final_length]
         if obj == "series":
             ser = pd.Series(data, index=index)
         else:

pandas/tests/extension/base/setitem.py

@@ -136,13 +136,13 @@ def test_setitem_iloc_scalar_mixed(self, data):
 
     def test_setitem_iloc_scalar_single(self, data):
         df = pd.DataFrame({"B": data})
-        df.iloc[10, 0] = data[1]
-        assert df.loc[10, "B"] == data[1]
+        df.iloc[9, 0] = data[1]
+        assert df.loc[9, "B"] == data[1]
 
     def test_setitem_iloc_scalar_multiple_homogoneous(self, data):
         df = pd.DataFrame({"A": data, "B": data})
-        df.iloc[10, 1] = data[1]
-        assert df.loc[10, "B"] == data[1]
+        df.iloc[9, 1] = data[1]
+        assert df.loc[9, "B"] == data[1]
 
     @pytest.mark.parametrize(
         "mask",
@@ -281,9 +281,9 @@ def test_setitem_mask_broadcast(self, data, setter):
         else:  # __setitem__
             target = ser
 
-        target[mask] = data[10]
-        assert ser[0] == data[10]
-        assert ser[1] == data[10]
+        target[mask] = data[9]
+        assert ser[0] == data[9]
+        assert ser[1] == data[9]
 
     def test_setitem_expand_columns(self, data):
         df = pd.DataFrame({"A": data})

pandas/tests/extension/json/test_json.py

@@ -190,9 +190,8 @@ def test_ffill_limit_area(
             )
 
     def test_value_counts(self, all_data, dropna, request):
-        if len(all_data) == 100 or dropna:
-            mark = pytest.mark.xfail(reason="unhashable")
-            request.applymarker(mark)
+        if len(all_data) == 10 or dropna:
+            request.applymarker(unhashable)
         super().test_value_counts(all_data, dropna)
 
     @unhashable