wForget commented on code in PR #5079:
URL: https://github.com/apache/kyuubi/pull/5079#discussion_r1270461196
##########
extensions/spark/kyuubi-extension-spark-3-3/src/main/scala/org/apache/kyuubi/sql/RebalanceBeforeWriting.scala:
##########
@@ -27,7 +27,18 @@ trait RepartitionBuilderWithRebalance extends
RepartitionBuilder {
query: LogicalPlan): LogicalPlan = {
if (!conf.getConf(KyuubiSQLConf.INFER_REBALANCE_AND_SORT_ORDERS) ||
dynamicPartitionColumns.nonEmpty) {
- RebalancePartitions(dynamicPartitionColumns, query)
+ // TODO: make configurable
+ val constantConditions = InferDynamicPartitionConstantConditions
+ .infer(dynamicPartitionColumns, query)
Review Comment:
> For this case, does not `rebalance(t)` better than `rebalance` ? Can you
talk more why `rebalance` is better ?
Makes sense, I was eager to avoid large partitions and ignored the original
intention of this optimization.
--
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]
