Add pylibcudf.null_mask.null_count

python/cudf/cudf/_lib/column.pxd

@@ -13,7 +13,6 @@ from pylibcudf.libcudf.column.column_view cimport (
 from pylibcudf.libcudf.types cimport size_type
 from rmm.librmm.device_buffer cimport device_buffer
 
-cdef dtype_from_lists_column_view(column_view cv)
 cdef dtype_from_column_view(column_view cv)
 
 cdef class Column:
@@ -42,5 +41,3 @@ cdef class Column:
 
     @staticmethod
     cdef Column from_column_view(column_view, object)
-
-    cdef size_type compute_null_count(self) except? 0

python/cudf/cudf/_lib/column.pyx

@@ -43,7 +43,6 @@ from pylibcudf.libcudf.column.column_factories cimport (
 )
 from pylibcudf.libcudf.column.column_view cimport column_view
 from pylibcudf.libcudf.lists.lists_column_view cimport lists_column_view
-from pylibcudf.libcudf.null_mask cimport null_count as cpp_null_count
 from pylibcudf.libcudf.scalar.scalar cimport scalar
 
 from cudf._lib.scalar cimport DeviceScalar
@@ -346,7 +345,15 @@ cdef class Column:
     @property
     def null_count(self):
         if self._null_count is None:
-            self._null_count = self.compute_null_count()
+            if not self.nullable or self.size == 0:
+                self._null_count = 0
+            else:
+                with acquire_spill_lock():
+                    self._null_count = pylibcudf.null_mask.null_count(
+                        self.base_mask.get_ptr(mode="read"),
+                        self.offset,
+                        self.offset + self.size
+                    )
         return self._null_count
 
     @property
@@ -410,18 +417,6 @@ cdef class Column:
         else:
             return other_col
 
-    cdef libcudf_types.size_type compute_null_count(self) except? 0:
-        with acquire_spill_lock():
-            if not self.nullable:
-                return 0
-            return cpp_null_count(
-                <libcudf_types.bitmask_type*><uintptr_t>(
-                    self.base_mask.get_ptr(mode="read")
-                ),
-                self.offset,
-                self.offset + self.size
-            )
-
     cdef mutable_column_view mutable_view(self) except *:
         if isinstance(self.dtype, cudf.CategoricalDtype):
             col = self.base_children[0]

python/pylibcudf/pylibcudf/null_mask.pxd

@@ -1,4 +1,4 @@
-# Copyright (c) 2024, NVIDIA CORPORATION.
+# Copyright (c) 2024-2025, NVIDIA CORPORATION.
 
 from pylibcudf.libcudf.types cimport mask_state, size_type
 
@@ -16,3 +16,5 @@ cpdef DeviceBuffer create_null_mask(size_type size, mask_state state = *)
 cpdef tuple bitmask_and(list columns)
 
 cpdef tuple bitmask_or(list columns)
+
+cpdef size_type null_count(Py_ssize_t bitmask, size_type start, size_type stop)

python/pylibcudf/pylibcudf/null_mask.pyi

@@ -12,3 +12,4 @@ def create_null_mask(
 ) -> DeviceBuffer: ...
 def bitmask_and(columns: list[Column]) -> tuple[DeviceBuffer, int]: ...
 def bitmask_or(columns: list[Column]) -> tuple[DeviceBuffer, int]: ...
+def null_count(bitmask: int, start: int, stop: int) -> int: ...

python/pylibcudf/pylibcudf/null_mask.pyx

@@ -1,10 +1,11 @@
-# Copyright (c) 2024, NVIDIA CORPORATION.
+# Copyright (c) 2024-2025, NVIDIA CORPORATION.
 
 from libcpp.memory cimport make_unique
 from libcpp.pair cimport pair
 from libcpp.utility cimport move
 from pylibcudf.libcudf cimport null_mask as cpp_null_mask
 from pylibcudf.libcudf.types cimport mask_state, size_type
+from pylibcudf.utils cimport int_to_bitmask_ptr
 
 from rmm.librmm.device_buffer cimport device_buffer
 from rmm.pylibrmm.device_buffer cimport DeviceBuffer
@@ -20,6 +21,7 @@ __all__ = [
     "bitmask_or",
     "copy_bitmask",
     "create_null_mask",
+    "null_count",
 ]
 
 cdef DeviceBuffer buffer_to_python(device_buffer buf):
@@ -148,3 +150,28 @@ cpdef tuple bitmask_or(list columns):
         c_result = cpp_null_mask.bitmask_or(c_table.view())
 
     return buffer_to_python(move(c_result.first)), c_result.second
+
+
+cpdef size_type null_count(Py_ssize_t bitmask, size_type start, size_type stop):
+    """Given a validity bitmask, counts the number of null elements.
+
+    For details, see :cpp:func:`null_count`.
+
+    Parameters
+    ----------
+    bitmask : int
+        Integer pointer to the bitmask.
+
+    start : int
+        Index of the first bit to count (inclusive).
+
+    stop : int
+        Index of the last bit to count (exclusive).
+
+    Returns
+    -------
+    int
+        The number of null elements in the specified range.
+    """
+    with nogil:
+        return cpp_null_mask.null_count(int_to_bitmask_ptr(bitmask), start, stop)