use attrs

Author: TomAugspurger

pandas/core/frame.py

@@ -5316,8 +5316,9 @@ def _arith_op(left, right):
             with np.errstate(all="ignore"):
                 res_values = _arith_op(this.values, other.values)
             new_data = dispatch_fill_zeros(func, this.values, other.values, res_values)
-        # XXX: pass them here.
-        return this._construct_result(new_data)
+        return this._construct_result(new_data).__finalize__(
+            (self, other), method="combine_frame"
+        )
 
     def _combine_match_index(self, other, func):
         # at this point we have `self.index.equals(other.index)`

pandas/core/generic.py

@@ -187,7 +187,7 @@ class NDFrame(PandasObject, SelectionMixin):
             "ix",
         ]
     )  # type: FrozenSet[str]
-    _metadata = ["allows_duplicate_labels"]  # type: List[str]
+    _metadata = []  # type: List[str]
     _is_copy = None
     _data = None  # type: BlockManager
 
@@ -224,12 +224,16 @@ def __init__(
         object.__setattr__(self, "_is_copy", None)
         object.__setattr__(self, "_data", data)
         object.__setattr__(self, "_item_cache", {})
-        object.__setattr__(self, "allows_duplicate_labels", allow_duplicate_labels)
         if attrs is None:
             attrs = {}
         else:
             attrs = dict(attrs)
+        # need to add it to the dict here, since NDFrame.__setattr__
+        # also calls NDFrame.__getattr__...
+        # attrs['allows_duplicate_labels'] = allow_duplicate_labels
         object.__setattr__(self, "_attrs", attrs)
+        object.__setattr__(self, "allows_duplicate_labels", allow_duplicate_labels)
+        # self.allows_duplicate_labels = allow_duplicate_labels
 
     def _init_mgr(self, mgr, axes=None, dtype=None, copy=False):
         """ passed a manager and a axes dict """
@@ -251,21 +255,6 @@ def _init_mgr(self, mgr, axes=None, dtype=None, copy=False):
     # ----------------------------------------------------------------------
 
     @property
-    def allows_duplicate_labels(self):
-        """
-        Whether this object allows duplicate labels.
-        """
-        return self._allows_duplicate_labels
-
-    @allows_duplicate_labels.setter
-    def allows_duplicate_labels(self, value: bool):
-        value = bool(value)
-        if not value:
-            for ax in self.axes:
-                ax._maybe_check_unique()
-
-        self._allows_duplicate_labels = value
-
     def attrs(self) -> Dict[Hashable, Any]:
         """
         Dictionary of global attributes on this object.
@@ -278,6 +267,22 @@ def attrs(self) -> Dict[Hashable, Any]:
     def attrs(self, value: Mapping[Hashable, Any]) -> None:
         self._attrs = dict(value)
 
+    @property
+    def allows_duplicate_labels(self) -> bool:
+        """
+        Whether this object allows duplicate labels.
+        """
+        return self.attrs["allows_duplicate_labels"]
+
+    @allows_duplicate_labels.setter
+    def allows_duplicate_labels(self, value: bool):
+        value = bool(value)
+        if not value:
+            for ax in self.axes:
+                ax._maybe_check_unique()
+
+        self.attrs["allows_duplicate_labels"] = value
+
     @property
     def is_copy(self):
         """
@@ -5249,6 +5254,7 @@ def pipe(self, func, *args, **kwargs):
     >>> s = pd.Series(range(3))
     >>> s
     0    0
+
     1    1
     2    2
     dtype: int64
@@ -5287,18 +5293,18 @@ def finalize_name(objs):
 
         duplicate_labels = "allows_duplicate_labels"
 
-        # import pdb; pdb.set_trace()
         if isinstance(other, NDFrame):
-            for name in other.attrs:
-                self.attrs[name] = other.attrs[name]
+            for name, value in other.attrs.items():
+                # Need to think about this...
+                if name == "allows_duplicate_labels":
+                    self.allows_duplicate_labels = value
+                elif name in self.attrs:
+                    self.attrs[name] = other.attrs[name]
+
             # For subclasses using _metadata.
             for name in self._metadata:
-                if name == "name" and getattr(other, "ndim", None) == 1:
-                    # Calling hasattr(other, 'name') is bad for DataFrames with
-                    # a name column.
-                    object.__setattr__(self, name, getattr(other, name, None))
-                elif name != "name":
-                    object.__setattr__(self, name, getattr(other, name, None))
+                object.__setattr__(self, name, getattr(other, name, None))
+
         elif method == "concat":
             assert isinstance(other, _Concatenator)
             self.allows_duplicate_labels = merge_all(other.objs, duplicate_labels)
@@ -5307,7 +5313,7 @@ def finalize_name(objs):
             self.allows_duplicate_labels = merge_all(
                 (other.left, other.right), duplicate_labels
             )
-        elif method in {"combine_const", "combine_frame"}:
+        elif method in {"combine_const", "combine_frame", "combine_series_frame"}:
             assert isinstance(other, tuple)
             self.allows_duplicate_labels = merge_all(other, duplicate_labels)
         elif method == "align_series":

pandas/core/groupby/generic.py

@@ -243,7 +243,7 @@ def aggregate(self, func=None, *args, **kwargs):
 
         if isinstance(func, str):
             return getattr(self, func)(*args, **kwargs).__finalize__(
-                method="groupby-aggregate"
+                self, method="groupby-aggregate"
             )
 
         if isinstance(func, abc.Iterable):
@@ -275,12 +275,16 @@ def aggregate(self, func=None, *args, **kwargs):
             print("Warning, ignoring as_index=True")
 
         # _level handled at higher
-        if not _level and isinstance(ret, dict):
+        if not _level and isinstance(ret, (dict, OrderedDict)):
             from pandas import concat
 
             ret = concat(ret, axis=1)
 
-        return ret.__finalize__(self, method="groupby-aggregate")
+        if isinstance(ret, NDFrame):
+            # TODO: when is this *not* an NDFrame?
+            # pandas/tests/resample/test_resample_api.py::test_agg_nested_dicts
+            ret = ret.__finalize__(self, method="groupby-aggregate")
+        return ret
 
     agg = aggregate
 
@@ -876,7 +880,7 @@ def aggregate(self, func=None, *args, **kwargs):
 
         result, how = self._aggregate(func, _level=_level, *args, **kwargs)
         if how is None:
-            return result
+            return result.__finalize__(self, method="groupby-aggregate")
 
         if result is None:
 

pandas/core/ops/__init__.py

@@ -624,7 +624,9 @@ def _combine_series_frame(self, other, func, fill_value=None, axis=None, level=N
     else:
         new_data = dispatch_to_series(left, right, func, axis="columns")
 
-    return left._construct_result(new_data)
+    return left._construct_result(new_data).__finalize__(
+        (self, other), method="combine_series_frame"
+    )
 
 
 def _align_method_FRAME(left, right, axis):
@@ -724,7 +726,7 @@ def f(self, other, axis=default_axis, level=None, fill_value=None):
                 self = self.fillna(fill_value)
 
             new_data = dispatch_to_series(self, other, op)
-            return self._construct_result(new_data)
+            return self._construct_result(new_data).__finalize__(self)
 
     f.__name__ = op_name
 

pandas/tests/test_duplicate_labels.py

@@ -249,7 +249,7 @@ class TestRaises:
     )
     def test_construction_with_duplicates(self, cls, axes):
         result = cls(**axes)
-        assert result._allows_duplicate_labels is True
+        assert result.allows_duplicate_labels is True
 
         with pytest.raises(pandas.errors.DuplicateLabelError):
             cls(**axes, allow_duplicate_labels=False)

pandas/util/testing.py

@@ -2800,7 +2800,7 @@ def inner(*args, **kwargs):
 
 
 class SubclassedSeries(Series):
-    _metadata = ["testattr", "name"]
+    _metadata = Series._metadata + ["testattr"]
 
     @property
     def _constructor(self):