Push DeviceScalar construction into cython for list.contains

python/cudf/cudf/_lib/lists.pyx

@@ -145,7 +145,10 @@ def extract_element(Column col, size_type index):
     return result
 
 
-def contains_scalar(Column col, DeviceScalar search_key):
+def contains_scalar(Column col, object py_search_key):
+
+    cdef DeviceScalar search_key = py_search_key.device_value
+
     cdef shared_ptr[lists_column_view] list_view = (
         make_shared[lists_column_view](col.view())
     )

python/cudf/cudf/core/column/lists.py

@@ -237,9 +237,10 @@ def contains(self, search_key):
         Series([False, True, True])
         dtype: bool
         """
+        search_key = cudf.Scalar(search_key)
         try:
             res = self._return_or_inplace(
-                contains_scalar(self._column, search_key.device_value)
+                contains_scalar(self._column, search_key)
             )
         except RuntimeError as e:
             if (

python/cudf/cudf/tests/test_contains.py

@@ -1,11 +1,13 @@
 from datetime import datetime as dt
 
+import numpy as np
 import pandas as pd
 import pytest
 
 from cudf import Series
 from cudf.core.index import RangeIndex, as_index
 from cudf.tests.utils import assert_eq
+from cudf.utils.dtypes import SCALAR_TYPES
 
 
 def cudf_date_series(start, stop, freq):
@@ -72,3 +74,21 @@ def test_index_contains(values, item, expected):
 def test_rangeindex_contains():
     assert_eq(True, 9 in RangeIndex(start=0, stop=10, name="Index"))
     assert_eq(False, 10 in RangeIndex(start=0, stop=10, name="Index"))
+
+
+@pytest.mark.parametrize("dtype", SCALAR_TYPES)
+def test_lists_contains(dtype):
+    dtype = np.dtype(dtype)
+    inner_data = np.array([1, 2, 3], dtype=dtype)
+    data = Series([inner_data])
+
+    if dtype.kind in "mM":
+        unit, _ = np.datetime_data(dtype)
+        contained_scalar = inner_data.dtype.type(2, unit)
+        not_contained_scalar = inner_data.dtype.type(42, unit)
+    else:
+        contained_scalar = inner_data.dtype.type(2)
+        not_contained_scalar = inner_data.dtype.type(42)
+
+    assert data.list.contains(contained_scalar)[0]
+    assert not data.list.contains(not_contained_scalar)[0]

python/cudf/cudf/utils/dtypes.py

@@ -103,6 +103,13 @@
 STRING_TYPES = {"object"}
 BOOL_TYPES = {"bool"}
 ALL_TYPES = NUMERIC_TYPES | DATETIME_TYPES | TIMEDELTA_TYPES | OTHER_TYPES
+SCALAR_TYPES = (
+    NUMERIC_TYPES
+    | BOOL_TYPES
+    | STRING_TYPES
+    | TIMEDELTA_TYPES
+    | DATETIME_TYPES
+)
 
 
 def np_to_pa_dtype(dtype):