working?

Author: MarcoGorelli

pandas/core/interchange/buffer.py

@@ -83,50 +83,51 @@ class PandasPyarrowBackedBuffer(Buffer):
     Data in the buffer is guaranteed to be contiguous in memory.
     """
 
-    def __init__(self, x: Any, allow_copy: bool = True) -> None:
+    def __init__(self, arr: Any, allow_copy: bool = True) -> None:
         """
         Handle only regular columns (= numpy arrays) for now.
         """
 
         # Store the numpy array in which the data resides as a private
         # attribute, so we can use it to retrieve the public attributes
-        self._x = x
+        self._buffer = arr.chunks[0].buffers()[1]
+        self._length = len(arr)
 
     @property
     def bufsize(self) -> int:
         """
         Buffer size in bytes.
         """
-        return self._x.size
+        return self._buffer.size * self._length
 
     @property
     def ptr(self) -> int:
         """
         Pointer to start of the buffer as an integer.
         """
-        return self._x.address
+        return self._buffer.address
 
     def __dlpack__(self) -> Any:
         """
         Represent this structure as DLPack interface.
         """
-        return self._x.__dlpack__()
+        return self._buffer.__dlpack__()
 
     def __dlpack_device__(self) -> tuple[DlpackDeviceType, int | None]:
         """
         Device type and device ID for where the data in the buffer resides.
         """
         return (DlpackDeviceType.CPU, None)
 
-    # def __repr__(self) -> str:
-    #     return (
-    #         "PandasBuffer("
-    #         + str(
-    #             {
-    #                 "bufsize": self.bufsize,
-    #                 "ptr": self.ptr,
-    #                 "device": self.__dlpack_device__()[0].name,
-    #             }
-    #         )
-    #         + ")"
-    #     )
+    def __repr__(self) -> str:
+        return (
+            "PandasBuffer("
+            + str(
+                {
+                    "bufsize": self.bufsize,
+                    "ptr": self.ptr,
+                    "device": self.__dlpack_device__()[0].name,
+                }
+            )
+            + ")"
+        )

pandas/core/interchange/column.py

@@ -199,11 +199,7 @@ def describe_null(self):
             return column_null_dtype, null_value
         kind = self.dtype[0]
         if isinstance(self._col.dtype, ArrowDtype):
-            column_null_dtype = ColumnNullType.USE_BITMASK
-            null_value = 0
-            if ~self._col.isna().any():
-                return column_null_dtype, None
-            return column_null_dtype, null_value
+            return ColumnNullType.USE_BITMASK, 0
         try:
             null, value = _NULL_DESCRIPTION[kind]
         except KeyError as err:
@@ -315,7 +311,7 @@ def _get_data_buffer(
             arr = self._col.array
             if isinstance(self._col.dtype, ArrowDtype):
                 arr = self._col.array
-                buffer = PandasPyarrowBackedBuffer(arr._pa_array.chunks[0].buffers()[1])
+                buffer = PandasPyarrowBackedBuffer(arr._pa_array)
                 return buffer, dtype
             if isinstance(self._col.dtype, BaseMaskedDtype):
                 np_arr = arr._data  # type: ignore[attr-defined]
@@ -368,7 +364,7 @@ def _get_validity_buffer(self) -> tuple[PandasBuffer, Any]:
             dtype = (DtypeKind.BOOL, 1, ArrowCTypes.BOOL, Endianness.NATIVE)
             if buf is None:
                 return buf, dtype
-            buffer = PandasPyarrowBackedBuffer(buf)
+            buffer = PandasPyarrowBackedBuffer(arr._pa_array)
             return buffer, dtype
 
         if isinstance(self._col.dtype, BaseMaskedDtype):