kgyrtkirk commented on a change in pull request #2360:
URL: https://github.com/apache/hive/pull/2360#discussion_r664637117
##########
File path: ql/src/java/org/apache/hadoop/hive/ql/optimizer/StatsOptimizer.java
##########
@@ -397,6 +397,11 @@ else if (getGbyKeyType(cgbyOp) == GbyKeyType.CONSTANT &&
rowCnt == 0) {
if (udaf instanceof GenericUDAFSum) {
// long/double/decimal
ExprNodeDesc desc = aggr.getParameters().get(0);
+ // return null for SUM(1), when the table is empty. Without this,
category = LONG, and the result is 0
+ // instead of NULL.
+ if (desc instanceof ExprNodeConstantDesc && rowCnt == 0) {
+ return oneRow.add(null);
Review comment:
this `return` doesn't look right to me - it may work correctly if you
around only selecting the aggregate - but what happens if you have more?
I believe you should execute `oneRow.add(null)` ; bail out the rest of this
conditional - and let the code continue with the remaining aggregates if there
is any
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
