Raise error when mixed types are being constructed (#13816)

This PR raises error when a mixed type data is being constructed instead of type-casting `nan` values to string nans (`'nan'`)

Authors:
  - GALI PREM SAGAR (https://github.com/galipremsagar)

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

URL: #13816

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

@@ -2161,9 +2161,7 @@ def as_column(
             if dtype is not None:
                 data = data.astype(dtype)
         elif arb_dtype.kind in ("O", "U"):
-            data = as_column(
-                pa.Array.from_pandas(arbitrary), dtype=arbitrary.dtype
-            )
+            data = as_column(pa.array(arbitrary), dtype=arbitrary.dtype)
             # There is no cast operation available for pa.Array from int to
             # str, Hence instead of handling in pa.Array block, we
             # will have to type-cast here.
@@ -2422,7 +2420,7 @@ def _construct_array(
         if (
             dtype is None
             and not cudf._lib.scalar._is_null_host_scalar(arbitrary)
-            and infer_dtype(arbitrary)
+            and infer_dtype(arbitrary, skipna=False)
             in (
                 "mixed",
                 "mixed-integer",

python/cudf/cudf/tests/test_series.py

@@ -2219,3 +2219,8 @@ def __getitem__(self, key):
 
     with pytest.raises(TypeError):
         cudf.Series(A())
+
+
+def test_series_constructor_error_mixed_type():
+    with pytest.raises(pa.ArrowTypeError):
+        cudf.Series(["abc", np.nan, "123"], nan_as_null=False)