Remove overflow error during decimal binops (#12063)

Fixes: #11337 

- [x] This PR removes raising of an overflow error and rather let's the data overflow similar to what we do with other numeric dtypes.

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

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

URL: #12063

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

@@ -399,4 +399,12 @@ def _get_decimal_type(lhs_dtype, rhs_dtype, op):
                 # to try the next dtype
                 continue
 
-    raise OverflowError("Maximum supported decimal type is Decimal128")
+    # Instead of raising an overflow error, we create a `Decimal128Dtype`
+    # with max possible scale & precision, see example of this demonstration
+    # here: https://learn.microsoft.com/en-us/sql/t-sql/data-types/
+    # precision-scale-and-length-transact-sql?view=sql-server-ver16#examples
+    scale = min(
+        scale, cudf.Decimal128Dtype.MAX_PRECISION - (precision - scale)
+    )
+    precision = min(cudf.Decimal128Dtype.MAX_PRECISION, max_precision)
+    return cudf.Decimal128Dtype(precision=precision, scale=scale)

python/cudf/cudf/tests/test_decimal.py

@@ -1,4 +1,4 @@
-# Copyright (c) 2021, NVIDIA CORPORATION.
+# Copyright (c) 2021-2022, NVIDIA CORPORATION.
 
 import decimal
 from decimal import Decimal
@@ -377,3 +377,9 @@ def test_decimal_invalid_precision():
 
     with pytest.raises(pa.ArrowInvalid):
         _ = cudf.Series([Decimal("300")], dtype=cudf.Decimal64Dtype(2, 1))
+
+
+def test_decimal_overflow():
+    s = cudf.Series([Decimal("0.0009384233522166997927180531650178250")])
+    result = s * s
+    assert_eq(cudf.Decimal128Dtype(precision=38, scale=37), result.dtype)