API: dont do inference on object-dtype arithmetic results

doc/source/whatsnew/v2.0.0.rst

@@ -348,6 +348,7 @@ Other API changes
 - Passing a sequence containing a type that cannot be converted to :class:`Timedelta` to :func:`to_timedelta` or to the :class:`Series` or :class:`DataFrame` constructor with ``dtype="timedelta64[ns]"`` or to :class:`TimedeltaIndex` now raises ``TypeError`` instead of ``ValueError`` (:issue:`49525`)
 - Changed behavior of :class:`Index` constructor with sequence containing at least one ``NaT`` and everything else either ``None`` or ``NaN`` to infer ``datetime64[ns]`` dtype instead of ``object``, matching :class:`Series` behavior (:issue:`49340`)
 - :func:`read_stata` with parameter ``index_col`` set to ``None`` (the default) will now set the index on the returned :class:`DataFrame` to a :class:`RangeIndex` instead of a :class:`Int64Index` (:issue:`49745`)
+- Changed behavior of :class:`Index`, :class:`Series`, and :class:`DataFrame` arithmetic methods when working with object-dtypes, the results no longer do type inference on the result of the array operations (:issue:`49714`)
 - Changed behavior of :class:`Index` constructor with an object-dtype ``numpy.ndarray`` containing all-``bool`` values or all-complex values, this will now retain object dtype, consistent with the :class:`Series` behavior (:issue:`49594`)
 - Changed behavior of :meth:`DataFrame.shift` with ``axis=1``, an integer ``fill_value``, and homogeneous datetime-like dtype, this now fills new columns with integer dtypes instead of casting to datetimelike (:issue:`49842`)
 - :meth:`DataFrame.values`, :meth:`DataFrame.to_numpy`, :meth:`DataFrame.xs`, :meth:`DataFrame.reindex`, :meth:`DataFrame.fillna`, and :meth:`DataFrame.replace` no longer silently consolidate the underlying arrays; do ``df = df.copy()`` to ensure consolidation (:issue:`49356`)

pandas/core/indexes/base.py

@@ -6582,10 +6582,10 @@ def _logical_method(self, other, op):
     def _construct_result(self, result, name):
         if isinstance(result, tuple):
             return (
-                Index._with_infer(result[0], name=name),
-                Index._with_infer(result[1], name=name),
+                Index(result[0], name=name, dtype=result[0].dtype),
+                Index(result[1], name=name, dtype=result[1].dtype),
             )
-        return Index._with_infer(result, name=name)
+        return Index(result, name=name, dtype=result.dtype)
 
     def _arith_method(self, other, op):
         if (

pandas/core/ops/__init__.py

@@ -232,18 +232,27 @@ def align_method_FRAME(
 
     def to_series(right):
         msg = "Unable to coerce to Series, length must be {req_len}: given {given_len}"
+
+        # pass dtype to avoid doing inference, which would break consistency
+        #  with Index/Series ops
+        dtype = None
+        if getattr(right, "dtype", None) == object:
+            # can't pass right.dtype unconditionally as that would break on e.g.
+            #  datetime64[h] ndarray
+            dtype = object
+
         if axis is not None and left._get_axis_name(axis) == "index":
             if len(left.index) != len(right):
                 raise ValueError(
                     msg.format(req_len=len(left.index), given_len=len(right))
                 )
-            right = left._constructor_sliced(right, index=left.index)
+            right = left._constructor_sliced(right, index=left.index, dtype=dtype)
         else:
             if len(left.columns) != len(right):
                 raise ValueError(
                     msg.format(req_len=len(left.columns), given_len=len(right))
                 )
-            right = left._constructor_sliced(right, index=left.columns)
+            right = left._constructor_sliced(right, index=left.columns, dtype=dtype)
         return right
 
     if isinstance(right, np.ndarray):
@@ -252,13 +261,25 @@ def to_series(right):
             right = to_series(right)
 
         elif right.ndim == 2:
+            # We need to pass dtype=right.dtype to retain object dtype
+            #  otherwise we lose consistency with Index and array ops
+            dtype = None
+            if getattr(right, "dtype", None) == object:
+                # can't pass right.dtype unconditionally as that would break on e.g.
+                #  datetime64[h] ndarray
+                dtype = object
+
             if right.shape == left.shape:
-                right = left._constructor(right, index=left.index, columns=left.columns)
+                right = left._constructor(
+                    right, index=left.index, columns=left.columns, dtype=dtype
+                )
 
             elif right.shape[0] == left.shape[0] and right.shape[1] == 1:
                 # Broadcast across columns
                 right = np.broadcast_to(right, left.shape)
-                right = left._constructor(right, index=left.index, columns=left.columns)
+                right = left._constructor(
+                    right, index=left.index, columns=left.columns, dtype=dtype
+                )
 
             elif right.shape[1] == left.shape[1] and right.shape[0] == 1:
                 # Broadcast along rows
@@ -411,7 +432,10 @@ def _maybe_align_series_as_frame(frame: DataFrame, series: Series, axis: AxisInt
         rvalues = rvalues.reshape(1, -1)
 
     rvalues = np.broadcast_to(rvalues, frame.shape)
-    return type(frame)(rvalues, index=frame.index, columns=frame.columns)
+    # pass dtype to avoid doing inference
+    return type(frame)(
+        rvalues, index=frame.index, columns=frame.columns, dtype=rvalues.dtype
+    )
 
 
 def flex_arith_method_FRAME(op):

pandas/core/series.py

@@ -2995,9 +2995,10 @@ def _construct_result(
             assert isinstance(res2, Series)
             return (res1, res2)
 
-        # We do not pass dtype to ensure that the Series constructor
-        #  does inference in the case where `result` has object-dtype.
-        out = self._constructor(result, index=self.index)
+        # TODO: result should always be ArrayLike, but this fails for some
+        #  JSONArray tests
+        dtype = getattr(result, "dtype", None)
+        out = self._constructor(result, index=self.index, dtype=dtype)
         out = out.__finalize__(self)
 
         # Set the result's name after __finalize__ is called because __finalize__

pandas/tests/arithmetic/test_object.py

@@ -187,7 +187,8 @@ def test_series_with_dtype_radd_timedelta(self, dtype):
             dtype=dtype,
         )
         expected = Series(
-            [pd.Timedelta("4 days"), pd.Timedelta("5 days"), pd.Timedelta("6 days")]
+            [pd.Timedelta("4 days"), pd.Timedelta("5 days"), pd.Timedelta("6 days")],
+            dtype=dtype,
         )
 
         result = pd.Timedelta("3 days") + ser
@@ -227,7 +228,9 @@ def test_mixed_timezone_series_ops_object(self):
             name="xxx",
         )
         assert ser2.dtype == object
-        exp = Series([pd.Timedelta("2 days"), pd.Timedelta("4 days")], name="xxx")
+        exp = Series(
+            [pd.Timedelta("2 days"), pd.Timedelta("4 days")], name="xxx", dtype=object
+        )
         tm.assert_series_equal(ser2 - ser, exp)
         tm.assert_series_equal(ser - ser2, -exp)
 
@@ -238,7 +241,11 @@ def test_mixed_timezone_series_ops_object(self):
         )
         assert ser.dtype == object
 
-        exp = Series([pd.Timedelta("01:30:00"), pd.Timedelta("02:30:00")], name="xxx")
+        exp = Series(
+            [pd.Timedelta("01:30:00"), pd.Timedelta("02:30:00")],
+            name="xxx",
+            dtype=object,
+        )
         tm.assert_series_equal(ser + pd.Timedelta("00:30:00"), exp)
         tm.assert_series_equal(pd.Timedelta("00:30:00") + ser, exp)
 

pandas/tests/arithmetic/test_timedelta64.py

@@ -1398,7 +1398,7 @@ def test_td64arr_addsub_anchored_offset_arraylike(self, obox, box_with_array):
     # ------------------------------------------------------------------
     # Unsorted
 
-    def test_td64arr_add_sub_object_array(self, box_with_array):
+    def test_td64arr_add_sub_object_array(self, box_with_array, using_array_manager):
         box = box_with_array
         xbox = np.ndarray if box is pd.array else box
 
@@ -1414,6 +1414,11 @@ def test_td64arr_add_sub_object_array(self, box_with_array):
             [Timedelta(days=2), Timedelta(days=4), Timestamp("2000-01-07")]
         )
         expected = tm.box_expected(expected, xbox)
+        if not using_array_manager:
+            # TODO: avoid mismatched behavior. This occurs bc inference
+            #  can happen within TimedeltaArray method, which means results
+            #  depend on whether we split blocks.
+            expected = expected.astype(object)
         tm.assert_equal(result, expected)
 
         msg = "unsupported operand type|cannot subtract a datelike"
@@ -1426,6 +1431,8 @@ def test_td64arr_add_sub_object_array(self, box_with_array):
 
         expected = pd.Index([Timedelta(0), Timedelta(0), Timestamp("2000-01-01")])
         expected = tm.box_expected(expected, xbox)
+        if not using_array_manager:
+            expected = expected.astype(object)
         tm.assert_equal(result, expected)