BUG: fix the return value of .all() and .any() on pd.Series

pandas/core/nanops.py

@@ -421,8 +421,27 @@ def nanany(values, axis=None, skipna: bool = True, mask=None):
     >>> nanops.nanany(s)
     False
     """
-    values, _, _, _, _ = _get_values(values, skipna, fill_value=False, mask=mask)
-    return values.any(axis)
+    values, _, dtype, _, _ = _get_values(values, skipna, fill_value=False, mask=mask)
+
+    # GH #12863
+    # Checking if the `axis` is None because numpy
+    # doesn't handle ``any`` and ``all`` on
+    # object arrays correctly. see
+    # https://github.com/numpy/numpy/issues/4352
+
+    # TODO: Find a less code-smelly way of doing this
+    if is_object_dtype(dtype) and axis is None:
+        output = values.any()
+    else:
+        output = values.any(axis)
+
+    if isinstance(output, bool):
+        return output
+
+    try:
+        return any(values)
+    except ValueError:
+        return values.any()
 
 
 def nanall(values, axis=None, skipna: bool = True, mask=None):
@@ -453,8 +472,27 @@ def nanall(values, axis=None, skipna: bool = True, mask=None):
     >>> nanops.nanall(s)
     False
     """
-    values, _, _, _, _ = _get_values(values, skipna, fill_value=True, mask=mask)
-    return values.all(axis)
+    values, _, dtype, _, _ = _get_values(values, skipna, fill_value=True, mask=mask)
+
+    # GH #12863
+    # Checking if the `axis` is None because numpy
+    # doesn't handle ``any`` and ``all`` on
+    # object arrays correctly. see
+    # https://github.com/numpy/numpy/issues/4352
+
+    # TODO: Find a less code-smelly way of doing this
+    if is_object_dtype(dtype) and axis is None:
+        output = values.all()
+    else:
+        output = values.all(axis)
+
+    if isinstance(output, bool):
+        return output
+
+    try:
+        return all(values)
+    except ValueError:
+        return values.all()
 
 
 @disallow("M8")

pandas/tests/reductions/test_reductions.py

@@ -811,19 +811,24 @@ def test_all_any(self):
         assert not bool_series.all()
         assert bool_series.any()
 
-        # Alternative types, with implicit 'object' dtype.
-        s = Series(["abc", True])
-        assert "abc" == s.any()  # 'abc' || True => 'abc'
-
     def test_all_any_params(self):
         # Check skipna, with implicit 'object' dtype.
         s1 = Series([np.nan, True])
         s2 = Series([np.nan, False])
-        assert s1.all(skipna=False)  # nan && True => True
+
+        # GH #12863
         assert s1.all(skipna=True)
-        assert np.isnan(s2.any(skipna=False))  # nan || False => nan
+        assert s1.any(skipna=True)
+
+        assert s1.all(skipna=False)
+        assert s1.any(skipna=False)
+
+        assert not s2.all(skipna=True)
         assert not s2.any(skipna=True)
 
+        assert not s2.all(skipna=False)
+        assert s2.any(skipna=False)
+
         # Check level.
         s = pd.Series([False, False, True, True, False, True], index=[0, 0, 1, 1, 2, 2])
         tm.assert_series_equal(s.all(level=0), Series([False, True, False]))
@@ -841,6 +846,18 @@ def test_all_any_params(self):
         with pytest.raises(NotImplementedError):
             s.all(bool_only=True)
 
+    def test_all_any_object_dtype(self):
+        # GH 12863
+
+        s1 = Series(["abc", True])
+        s2 = Series(["abc", False])
+
+        assert s1.all()
+        assert s1.any()
+
+        assert not s2.all()
+        assert s2.any()
+
     def test_timedelta64_analytics(self):
 
         # index min/max

pandas/tests/series/test_apply.py

@@ -390,8 +390,8 @@ def test_non_callable_aggregates(self):
                     ("sum", "abc"),
                     ("max", "c"),
                     ("min", "a"),
-                    ("all", "c"),  # see GH12863
-                    ("any", "a"),
+                    ("all", True),
+                    ("any", True),
                 ],
             ),
         ),

pandas/tests/test_nanops.py

@@ -270,6 +270,10 @@ def _badobj_wrap(self, value, func, allow_complex=True, **kwargs):
                 value = value.astype("f8")
         return func(value, **kwargs)
 
+    # GH #12863
+    # Disabled until https://github.com/numpy/numpy/issues/4352
+    # is fixed
+    @pytest.mark.xfail(reason="disabled")
     @pytest.mark.parametrize(
         "nan_op,np_op", [(nanops.nanany, np.any), (nanops.nanall, np.all)]
     )