BUG: Preserve subclassing with groupby operations

doc/source/whatsnew/v0.25.2.rst

@@ -78,7 +78,7 @@ Groupby/resample/rolling
 
 - Bug incorrectly raising an ``IndexError`` when passing a list of quantiles to :meth:`pandas.core.groupby.DataFrameGroupBy.quantile` (:issue:`28113`).
 - Bug in :meth:`pandas.core.groupby.GroupBy.shift`, :meth:`pandas.core.groupby.GroupBy.bfill` and :meth:`pandas.core.groupby.GroupBy.ffill` where timezone information would be dropped (:issue:`19995`, :issue:`27992`)
--
+- Bug groupby/resampling where subclasses were not returned from groupby ops (:issue:`28330`)
 -
 -
 

pandas/core/groupby/generic.py

@@ -337,7 +337,7 @@ def _aggregate_item_by_item(self, func, *args, **kwargs):
             if not len(result_columns) and errors is not None:
                 raise errors
 
-        return DataFrame(result, columns=result_columns)
+        return self.obj._constructor(result, columns=result_columns)
 
     def _decide_output_index(self, output, labels):
         if len(output) == len(labels):
@@ -356,7 +356,7 @@ def _decide_output_index(self, output, labels):
 
     def _wrap_applied_output(self, keys, values, not_indexed_same=False):
         if len(keys) == 0:
-            return DataFrame(index=keys)
+            return self.obj._constructor(index=keys)
 
         key_names = self.grouper.names
 
@@ -372,7 +372,7 @@ def first_not_none(values):
         if v is None:
             # GH9684. If all values are None, then this will throw an error.
             # We'd prefer it return an empty dataframe.
-            return DataFrame()
+            return self.obj._constructor()
         elif isinstance(v, DataFrame):
             return self._concat_objects(keys, values, not_indexed_same=not_indexed_same)
         elif self.grouper.groupings is not None:
@@ -401,7 +401,7 @@ def first_not_none(values):
             # make Nones an empty object
             v = first_not_none(values)
             if v is None:
-                return DataFrame()
+                return self.obj._constructor()
             elif isinstance(v, NDFrame):
                 values = [
                     x if x is not None else v._constructor(**v._construct_axes_dict())
@@ -467,7 +467,7 @@ def first_not_none(values):
                             or isinstance(key_index, MultiIndex)
                         ):
                             stacked_values = np.vstack([np.asarray(v) for v in values])
-                            result = DataFrame(
+                            result = self.obj._constructor(
                                 stacked_values, index=key_index, columns=index
                             )
                         else:
@@ -484,14 +484,16 @@ def first_not_none(values):
                             result.columns = index
                     else:
                         stacked_values = np.vstack([np.asarray(v) for v in values])
-                        result = DataFrame(
+                        result = self.obj._constructor(
                             stacked_values.T, index=v.index, columns=key_index
                         )
 
                 except (ValueError, AttributeError):
                     # GH1738: values is list of arrays of unequal lengths fall
                     # through to the outer else caluse
-                    return Series(values, index=key_index, name=self._selection_name)
+                    return self.obj._constructor_sliced(
+                        values, index=key_index, name=self._selection_name
+                    )
 
                 # if we have date/time like in the original, then coerce dates
                 # as we are stacking can easily have object dtypes here
@@ -510,7 +512,7 @@ def first_not_none(values):
                 # self._selection_name not passed through to Series as the
                 # result should not take the name of original selection
                 # of columns
-                return Series(values, index=key_index)._convert(
+                return self.obj._constructor_sliced(values, index=key_index)._convert(
                     datetime=True, coerce=coerce
                 )
 
@@ -554,7 +556,7 @@ def _transform_general(self, func, *args, **kwargs):
                     r.columns = group.columns
                     r.index = group.index
                 else:
-                    r = DataFrame(
+                    r = self.obj._constructor(
                         np.concatenate([res.values] * len(group.index)).reshape(
                             group.shape
                         ),
@@ -681,7 +683,7 @@ def _transform_item_by_item(self, obj, wrapper):
         if len(output) < len(obj.columns):
             columns = columns.take(inds)
 
-        return DataFrame(output, index=obj.index, columns=columns)
+        return self.obj._constructor(output, index=obj.index, columns=columns)
 
     def filter(self, func, dropna=True, *args, **kwargs):
         """
@@ -706,7 +708,7 @@ def filter(self, func, dropna=True, *args, **kwargs):
 
         Examples
         --------
-        >>> df = pd.DataFrame({'A' : ['foo', 'bar', 'foo', 'bar',
+        >>> df = pd.self.obj._constructor({'A' : ['foo', 'bar', 'foo', 'bar',
         ...                           'foo', 'bar'],
         ...                    'B' : [1, 2, 3, 4, 5, 6],
         ...                    'C' : [2.0, 5., 8., 1., 2., 9.]})
@@ -875,7 +877,7 @@ def aggregate(self, func=None, *args, **kwargs):
                 result = self._aggregate_named(func, *args, **kwargs)
 
             index = Index(sorted(result), name=self.grouper.names[0])
-            ret = Series(result, index=index)
+            ret = self.obj._constructor(result, index=index)
 
         if not self.as_index:  # pragma: no cover
             print("Warning, ignoring as_index=True")
@@ -943,20 +945,21 @@ def _aggregate_multiple_funcs(self, arg, _level):
             # let higher level handle
             if _level:
                 return results
+            self.obj._constructor(results, columns=columns)
 
-        return DataFrame(results, columns=columns)
+        return self.obj._constructor_expanddim(results, columns=columns)
 
     def _wrap_output(self, output, index, names=None):
         """ common agg/transform wrapping logic """
         output = output[self._selection_name]
 
         if names is not None:
-            return DataFrame(output, index=index, columns=names)
+            return self.obj._constructor_expanddim(output, index=index, columns=names)
         else:
             name = self._selection_name
             if name is None:
                 name = self._selected_obj.name
-            return Series(output, index=index, name=name)
+            return self.obj._constructor(output, index=index, name=name)
 
     def _wrap_aggregated_output(self, output, names=None):
         result = self._wrap_output(
@@ -970,7 +973,7 @@ def _wrap_transformed_output(self, output, names=None):
     def _wrap_applied_output(self, keys, values, not_indexed_same=False):
         if len(keys) == 0:
             # GH #6265
-            return Series([], name=self._selection_name, index=keys)
+            return self.obj._constructor([], name=self._selection_name, index=keys)
 
         def _get_index():
             if self.grouper.nkeys > 1:
@@ -982,7 +985,9 @@ def _get_index():
         if isinstance(values[0], dict):
             # GH #823 #24880
             index = _get_index()
-            result = self._reindex_output(DataFrame(values, index=index))
+            result = self._reindex_output(
+                self.obj._constructor_expanddim(values, index=index)
+            )
             # if self.observed is False,
             # keep all-NaN rows created while re-indexing
             result = result.stack(dropna=self.observed)
@@ -996,7 +1001,9 @@ def _get_index():
             return self._concat_objects(keys, values, not_indexed_same=not_indexed_same)
         else:
             # GH #6265 #24880
-            result = Series(data=values, index=_get_index(), name=self._selection_name)
+            result = self.obj._constructor(
+                data=values, index=_get_index(), name=self._selection_name
+            )
             return self._reindex_output(result)
 
     def _aggregate_named(self, func, *args, **kwargs):
@@ -1052,7 +1059,7 @@ def transform(self, func, *args, **kwargs):
 
             result = concat(results).sort_index()
         else:
-            result = Series()
+            result = self.obj._constructor()
 
         # we will only try to coerce the result type if
         # we have a numeric dtype, as these are *always* udfs
@@ -1078,7 +1085,7 @@ def _transform_fast(self, func, func_nm):
         out = algorithms.take_1d(func()._values, ids)
         if cast:
             out = self._try_cast(out, self.obj)
-        return Series(out, index=self.obj.index, name=self.obj.name)
+        return self.obj._constructor(out, index=self.obj.index, name=self.obj.name)
 
     def filter(self, func, dropna=True, *args, **kwargs):  # noqa
         """
@@ -1193,7 +1200,7 @@ def nunique(self, dropna=True):
             res, out = np.zeros(len(ri), dtype=out.dtype), res
             res[ids[idx]] = out
 
-        return Series(res, index=ri, name=self._selection_name)
+        return self.obj._constructor(res, index=ri, name=self._selection_name)
 
     @Appender(Series.describe.__doc__)
     def describe(self, **kwargs):
@@ -1233,7 +1240,7 @@ def value_counts(
         else:
 
             # lab is a Categorical with categories an IntervalIndex
-            lab = cut(Series(val), bins, include_lowest=True)
+            lab = cut(self.obj._constructor(val), bins, include_lowest=True)
             lev = lab.cat.categories
             lab = lev.take(lab.cat.codes)
             llab = lambda lab, inc: lab[inc]._multiindex.codes[-1]
@@ -1293,7 +1300,7 @@ def value_counts(
 
             if is_integer_dtype(out):
                 out = ensure_int64(out)
-            return Series(out, index=mi, name=self._selection_name)
+            return self.obj._constructor(out, index=mi, name=self._selection_name)
 
         # for compat. with libgroupby.value_counts need to ensure every
         # bin is present at every index level, null filled with zeros
@@ -1322,7 +1329,7 @@ def value_counts(
 
         if is_integer_dtype(out):
             out = ensure_int64(out)
-        return Series(out, index=mi, name=self._selection_name)
+        return self.obj._constructor(out, index=mi, name=self._selection_name)
 
     def count(self):
         """
@@ -1341,7 +1348,7 @@ def count(self):
         minlength = ngroups or 0
         out = np.bincount(ids[mask], minlength=minlength)
 
-        return Series(
+        return self.obj._constructor(
             out,
             index=self.grouper.result_index,
             name=self._selection_name,
@@ -1513,9 +1520,11 @@ def _wrap_generic_output(self, result, obj):
         result_index = self.grouper.levels[0]
 
         if self.axis == 0:
-            return DataFrame(result, index=obj.columns, columns=result_index).T
+            return self.obj._constructor(
+                result, index=obj.columns, columns=result_index
+            ).T
         else:
-            return DataFrame(result, index=obj.index, columns=result_index)
+            return self.obj._constructor(result, index=obj.index, columns=result_index)
 
     def _get_data_to_aggregate(self):
         obj = self._obj_with_exclusions
@@ -1548,33 +1557,33 @@ def _wrap_aggregated_output(self, output, names=None):
         output_keys = self._decide_output_index(output, agg_labels)
 
         if not self.as_index:
-            result = DataFrame(output, columns=output_keys)
+            result = self.obj._constructor(output, columns=output_keys)
             self._insert_inaxis_grouper_inplace(result)
             result = result._consolidate()
         else:
             index = self.grouper.result_index
-            result = DataFrame(output, index=index, columns=output_keys)
+            result = self.obj._constructor(output, index=index, columns=output_keys)
 
         if self.axis == 1:
             result = result.T
 
         return self._reindex_output(result)._convert(datetime=True)
 
     def _wrap_transformed_output(self, output, names=None):
-        return DataFrame(output, index=self.obj.index)
+        return self.obj._constructor(output, index=self.obj.index)
 
     def _wrap_agged_blocks(self, items, blocks):
         if not self.as_index:
             index = np.arange(blocks[0].values.shape[-1])
             mgr = BlockManager(blocks, [items, index])
-            result = DataFrame(mgr)
+            result = self.obj._constructor(mgr)
 
             self._insert_inaxis_grouper_inplace(result)
             result = result._consolidate()
         else:
             index = self.grouper.result_index
             mgr = BlockManager(blocks, [items, index])
-            result = DataFrame(mgr)
+            result = self.obj._constructor(mgr)
 
         if self.axis == 1:
             result = result.T

pandas/tests/groupby/test_groupby.py

@@ -134,6 +134,34 @@ def func(dataf):
     result = df.groupby("X", squeeze=False).count()
     assert isinstance(result, DataFrame)
 
+    # https://github.com/pandas-dev/pandas/issues/28330
+    # Test groupby operations on subclassed dataframes/series
+    class ChildSeries(Series):
+        pass
+
+    class ChildDataFrame(DataFrame):
+        @property
+        def _constructor(self):
+            return ChildDataFrame
+
+        _constructor_sliced = ChildSeries
+
+    ChildSeries._constructor_expanddim = ChildDataFrame
+
+    cdf = ChildDataFrame(
+        [
+            {"val1": 1, "val2": 20},
+            {"val1": 1, "val2": 19},
+            {"val1": 2, "val2": 27},
+            {"val1": 2, "val2": 12},
+        ]
+    )
+    result = cdf.groupby("val1").sum()
+    assert isinstance(result, ChildDataFrame)
+    assert isinstance(result, DataFrame)
+    assert isinstance(result["val2"], ChildSeries)
+    assert isinstance(result["val2"], Series)
+
 
 def test_inconsistent_return_type():
     # GH5592