REF: remove unnecessary Block specialization (#54568)

* REF: remove unnecessary Block specialization

* revert accidental

* mypy fixup

* mypy fixup

Author: jbrockmendel

pandas/_libs/internals.pyi

@@ -15,7 +15,6 @@ from pandas._typing import (
 )
 
 from pandas import Index
-from pandas.core.arrays._mixins import NDArrayBackedExtensionArray
 from pandas.core.internals.blocks import Block as B
 
 def slice_len(slc: slice, objlen: int = ...) -> int: ...
@@ -60,7 +59,7 @@ class BlockPlacement:
     def append(self, others: list[BlockPlacement]) -> BlockPlacement: ...
     def tile_for_unstack(self, factor: int) -> npt.NDArray[np.intp]: ...
 
-class SharedBlock:
+class Block:
     _mgr_locs: BlockPlacement
     ndim: int
     values: ArrayLike
@@ -72,19 +71,8 @@ class SharedBlock:
         ndim: int,
         refs: BlockValuesRefs | None = ...,
     ) -> None: ...
-
-class NumpyBlock(SharedBlock):
-    values: np.ndarray
-    @final
-    def slice_block_rows(self, slicer: slice) -> Self: ...
-
-class NDArrayBackedBlock(SharedBlock):
-    values: NDArrayBackedExtensionArray
-    @final
     def slice_block_rows(self, slicer: slice) -> Self: ...
 
-class Block(SharedBlock): ...
-
 class BlockManager:
     blocks: tuple[B, ...]
     axes: list[Index]
@@ -100,7 +88,7 @@ class BlockManager:
 
 class BlockValuesRefs:
     referenced_blocks: list[weakref.ref]
-    def __init__(self, blk: SharedBlock | None = ...) -> None: ...
-    def add_reference(self, blk: SharedBlock) -> None: ...
+    def __init__(self, blk: Block | None = ...) -> None: ...
+    def add_reference(self, blk: Block) -> None: ...
     def add_index_reference(self, index: Index) -> None: ...
     def has_reference(self) -> bool: ...

pandas/_libs/internals.pyx

@@ -24,7 +24,6 @@ cnp.import_array()
 
 from pandas._libs.algos import ensure_int64
 
-from pandas._libs.arrays cimport NDArrayBacked
 from pandas._libs.util cimport (
     is_array,
     is_integer_object,
@@ -639,14 +638,19 @@ def _unpickle_block(values, placement, ndim):
 
 
 @cython.freelist(64)
-cdef class SharedBlock:
+cdef class Block:
     """
     Defining __init__ in a cython class significantly improves performance.
     """
     cdef:
         public BlockPlacement _mgr_locs
         public BlockValuesRefs refs
         readonly int ndim
+        # 2023-08-15 no apparent performance improvement from declaring values
+        #  as ndarray in a type-special subclass (similar for NDArrayBacked).
+        #  This might change if slice_block_rows can be optimized with something
+        #  like https://github.com/numpy/numpy/issues/23934
+        public object values
 
     def __cinit__(
         self,
@@ -666,6 +670,8 @@ cdef class SharedBlock:
         refs: BlockValuesRefs, optional
             Ref tracking object or None if block does not have any refs.
         """
+        self.values = values
+
         self._mgr_locs = placement
         self.ndim = ndim
         if refs is None:
@@ -699,51 +705,7 @@ cdef class SharedBlock:
             ndim = maybe_infer_ndim(self.values, self.mgr_locs)
             self.ndim = ndim
 
-
-cdef class NumpyBlock(SharedBlock):
-    cdef:
-        public ndarray values
-
-    def __cinit__(
-        self,
-        ndarray values,
-        BlockPlacement placement,
-        int ndim,
-        refs: BlockValuesRefs | None = None,
-    ):
-        # set values here; the (implicit) call to SharedBlock.__cinit__ will
-        #  set placement, ndim and refs
-        self.values = values
-
-    cpdef NumpyBlock slice_block_rows(self, slice slicer):
-        """
-        Perform __getitem__-like specialized to slicing along index.
-
-        Assumes self.ndim == 2
-        """
-        new_values = self.values[..., slicer]
-        return type(self)(new_values, self._mgr_locs, ndim=self.ndim, refs=self.refs)
-
-
-cdef class NDArrayBackedBlock(SharedBlock):
-    """
-    Block backed by NDArrayBackedExtensionArray
-    """
-    cdef public:
-        NDArrayBacked values
-
-    def __cinit__(
-        self,
-        NDArrayBacked values,
-        BlockPlacement placement,
-        int ndim,
-        refs: BlockValuesRefs | None = None,
-    ):
-        # set values here; the (implicit) call to SharedBlock.__cinit__ will
-        #  set placement, ndim and refs
-        self.values = values
-
-    cpdef NDArrayBackedBlock slice_block_rows(self, slice slicer):
+    cpdef Block slice_block_rows(self, slice slicer):
         """
         Perform __getitem__-like specialized to slicing along index.
 
@@ -753,22 +715,6 @@ cdef class NDArrayBackedBlock(SharedBlock):
         return type(self)(new_values, self._mgr_locs, ndim=self.ndim, refs=self.refs)
 
 
-cdef class Block(SharedBlock):
-    cdef:
-        public object values
-
-    def __cinit__(
-        self,
-        object values,
-        BlockPlacement placement,
-        int ndim,
-        refs: BlockValuesRefs | None = None,
-    ):
-        # set values here; the (implicit) call to SharedBlock.__cinit__ will
-        #  set placement, ndim and refs
-        self.values = values
-
-
 @cython.freelist(64)
 cdef class BlockManager:
     cdef:
@@ -811,7 +757,7 @@ cdef class BlockManager:
         cdef:
             intp_t blkno, i, j
             cnp.npy_intp length = self.shape[0]
-            SharedBlock blk
+            Block blk
             BlockPlacement bp
             ndarray[intp_t, ndim=1] new_blknos, new_blklocs
 
@@ -901,7 +847,7 @@ cdef class BlockManager:
 
     cdef BlockManager _slice_mgr_rows(self, slice slobj):
         cdef:
-            SharedBlock blk, nb
+            Block blk, nb
             BlockManager mgr
             ndarray blknos, blklocs
 
@@ -945,18 +891,18 @@ cdef class BlockValuesRefs:
     cdef:
         public list referenced_blocks
 
-    def __cinit__(self, blk: SharedBlock | None = None) -> None:
+    def __cinit__(self, blk: Block | None = None) -> None:
         if blk is not None:
             self.referenced_blocks = [weakref.ref(blk)]
         else:
             self.referenced_blocks = []
 
-    def add_reference(self, blk: SharedBlock) -> None:
+    def add_reference(self, blk: Block) -> None:
         """Adds a new reference to our reference collection.
 
         Parameters
         ----------
-        blk: SharedBlock
+        blk : Block
             The block that the new references should point to.
         """
         self.referenced_blocks.append(weakref.ref(blk))

pandas/core/generic.py

@@ -817,9 +817,7 @@ def swapaxes(self, axis1: Axis, axis2: Axis, copy: bool_t | None = None) -> Self
             assert isinstance(new_mgr, BlockManager)
             assert isinstance(self._mgr, BlockManager)
             new_mgr.blocks[0].refs = self._mgr.blocks[0].refs
-            new_mgr.blocks[0].refs.add_reference(
-                new_mgr.blocks[0]  # type: ignore[arg-type]
-            )
+            new_mgr.blocks[0].refs.add_reference(new_mgr.blocks[0])
             if not using_copy_on_write() and copy is not False:
                 new_mgr = new_mgr.copy(deep=True)
 

pandas/core/internals/blocks.py

@@ -147,7 +147,7 @@ def newfunc(self, *args, **kwargs) -> list[Block]:
     return cast(F, newfunc)
 
 
-class Block(PandasObject):
+class Block(PandasObject, libinternals.Block):
     """
     Canonical n-dimensional unit of homogeneous dtype contained in a pandas
     data structure
@@ -1936,7 +1936,7 @@ def pad_or_backfill(
         return [self.make_block_same_class(new_values)]
 
 
-class ExtensionBlock(libinternals.Block, EABackedBlock):
+class ExtensionBlock(EABackedBlock):
     """
     Block for holding extension types.
 
@@ -2204,7 +2204,7 @@ def _unstack(
         return blocks, mask
 
 
-class NumpyBlock(libinternals.NumpyBlock, Block):
+class NumpyBlock(Block):
     values: np.ndarray
     __slots__ = ()
 
@@ -2242,7 +2242,7 @@ class ObjectBlock(NumpyBlock):
     __slots__ = ()
 
 
-class NDArrayBackedExtensionBlock(libinternals.NDArrayBackedBlock, EABackedBlock):
+class NDArrayBackedExtensionBlock(EABackedBlock):
     """
     Block backed by an NDArrayBackedExtensionArray
     """

pandas/core/internals/managers.py

@@ -263,9 +263,7 @@ def add_references(self, mgr: BaseBlockManager) -> None:
             return
         for i, blk in enumerate(self.blocks):
             blk.refs = mgr.blocks[i].refs
-            # Argument 1 to "add_reference" of "BlockValuesRefs" has incompatible type
-            # "Block"; expected "SharedBlock"
-            blk.refs.add_reference(blk)  # type: ignore[arg-type]
+            blk.refs.add_reference(blk)
 
     def references_same_values(self, mgr: BaseBlockManager, blkno: int) -> bool:
         """

pandas/core/series.py

@@ -897,7 +897,7 @@ def view(self, dtype: Dtype | None = None) -> Series:
         if isinstance(res_ser._mgr, SingleBlockManager):
             blk = res_ser._mgr._block
             blk.refs = cast("BlockValuesRefs", self._references)
-            blk.refs.add_reference(blk)  # type: ignore[arg-type]
+            blk.refs.add_reference(blk)
         return res_ser.__finalize__(self, method="view")
 
     # ----------------------------------------------------------------------