viirya commented on code in PR #216:
URL:
https://github.com/apache/arrow-datafusion-comet/pull/216#discussion_r1532921316
##########
common/src/main/scala/org/apache/spark/sql/comet/util/Utils.scala:
##########
@@ -69,6 +69,7 @@ object Utils {
case int: ArrowType.Int if int.getIsSigned && int.getBitWidth == 8 * 2 =>
ShortType
case int: ArrowType.Int if int.getIsSigned && int.getBitWidth == 8 * 4 =>
IntegerType
case int: ArrowType.Int if int.getIsSigned && int.getBitWidth == 8 * 8 =>
LongType
+ case int: ArrowType.Int if int.getBitWidth == 8 * 8 => LongType
Review Comment:
If we have both partial and final aggregation operators in Comet, it should
be okay as Java doesn't process the intermediate results (state), but if we
have only partial aggregation in Comet, this Uint64 array as LongType will
possibly cause overflow in corner case, I think.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
