Make Frame.astype return Self instead of a ColumnAccessor (#15861)

Allows simplification for it's subclasses (`IndexFrame.astype`, `Index.astype`)

Also minor cleanups in the `equals` method

Authors:
  - Matthew Roeschke (https://github.com/mroeschke)

Approvers:
  - Bradley Dice (https://github.com/bdice)

URL: #15861

python/cudf/cudf/core/_base_index.py

@@ -149,7 +149,7 @@ def ndim(self) -> int:  # noqa: D401
         """Number of dimensions of the underlying data, by definition 1."""
         return 1
 
-    def equals(self, other):
+    def equals(self, other) -> bool:
         """
         Determine if two Index objects contain the same elements.
 

python/cudf/cudf/core/dataframe.py

@@ -2590,7 +2590,7 @@ def items(self):
             yield (k, self[k])
 
     @_cudf_nvtx_annotate
-    def equals(self, other):
+    def equals(self, other) -> bool:
         ret = super().equals(other)
         # If all other checks matched, validate names.
         if ret:

python/cudf/cudf/core/frame.py

@@ -273,20 +273,13 @@ def __len__(self) -> int:
         return self._num_rows
 
     @_cudf_nvtx_annotate
-    def astype(self, dtype, copy: bool = False):
-        result_data = {
-            col_name: col.astype(dtype.get(col_name, col.dtype), copy=copy)
+    def astype(self, dtype: dict[Any, Dtype], copy: bool = False) -> Self:
+        casted = (
+            col.astype(dtype.get(col_name, col.dtype), copy=copy)
             for col_name, col in self._data.items()
-        }
-
-        return ColumnAccessor(
-            data=result_data,
-            multiindex=self._data.multiindex,
-            level_names=self._data.level_names,
-            rangeindex=self._data.rangeindex,
-            label_dtype=self._data.label_dtype,
-            verify=False,
         )
+        ca = self._data._from_columns_like_self(casted, verify=False)
+        return self._from_data_like_self(ca)
 
     @_cudf_nvtx_annotate
     def equals(self, other) -> bool:
@@ -349,11 +342,7 @@ def equals(self, other) -> bool:
         """
         if self is other:
             return True
-        if (
-            other is None
-            or not isinstance(other, type(self))
-            or len(self) != len(other)
-        ):
+        if not isinstance(other, type(self)) or len(self) != len(other):
             return False
 
         return all(

python/cudf/cudf/core/index.py

@@ -445,7 +445,7 @@ def __getitem__(self, index):
         return self._as_int_index()[index]
 
     @_cudf_nvtx_annotate
-    def equals(self, other):
+    def equals(self, other) -> bool:
         if isinstance(other, RangeIndex):
             return self._range == other._range
         return self._as_int_index().equals(other)
@@ -1058,6 +1058,16 @@ def _from_data(cls, data: MutableMapping, name: Any = no_default) -> Self:
             out.name = name
         return out
 
+    @classmethod
+    @_cudf_nvtx_annotate
+    def _from_data_like_self(
+        cls, data: MutableMapping, name: Any = no_default
+    ) -> Self:
+        out = _index_from_data(data, name)
+        if name is not no_default:
+            out.name = name
+        return out
+
     @classmethod
     @_cudf_nvtx_annotate
     def from_arrow(cls, obj):
@@ -1180,12 +1190,8 @@ def is_unique(self):
         return self._column.is_unique
 
     @_cudf_nvtx_annotate
-    def equals(self, other):
-        if (
-            other is None
-            or not isinstance(other, BaseIndex)
-            or len(self) != len(other)
-        ):
+    def equals(self, other) -> bool:
+        if not isinstance(other, BaseIndex) or len(self) != len(other):
             return False
 
         check_dtypes = False
@@ -1231,7 +1237,7 @@ def copy(self, name=None, deep=False):
 
     @_cudf_nvtx_annotate
     def astype(self, dtype, copy: bool = True):
-        return _index_from_data(super().astype({self.name: dtype}, copy))
+        return super().astype({self.name: dtype}, copy)
 
     @_cudf_nvtx_annotate
     def get_indexer(self, target, method=None, limit=None, tolerance=None):

python/cudf/cudf/core/indexed_frame.py

@@ -625,10 +625,8 @@ def copy(self, deep: bool = True) -> Self:
         )
 
     @_cudf_nvtx_annotate
-    def equals(self, other):  # noqa: D102
-        if not super().equals(other):
-            return False
-        return self.index.equals(other.index)
+    def equals(self, other) -> bool:  # noqa: D102
+        return super().equals(other) and self.index.equals(other.index)
 
     @property
     def index(self):
@@ -4896,10 +4894,10 @@ def repeat(self, repeats, axis=None):
 
     def astype(
         self,
-        dtype,
+        dtype: dict[Any, Dtype],
         copy: bool = False,
         errors: Literal["raise", "ignore"] = "raise",
-    ):
+    ) -> Self:
         """Cast the object to the given dtype.
 
         Parameters
@@ -5010,14 +5008,12 @@ def astype(
             raise ValueError("invalid error value specified")
 
         try:
-            data = super().astype(dtype, copy)
+            return super().astype(dtype, copy)
         except Exception as e:
             if errors == "raise":
                 raise e
             return self
 
-        return self._from_data(data, index=self.index)
-
     @_cudf_nvtx_annotate
     def drop(
         self,