github-actions[bot] commented on code in PR #65657:
URL: https://github.com/apache/doris/pull/65657#discussion_r3600301740
##########
fe/fe-core/src/main/java/org/apache/doris/nereids/trees/plans/logical/LogicalOlapTableStreamScan.java:
##########
@@ -276,11 +276,11 @@ public LogicalOlapTableStreamScan
withSelectedPartitionIds(List<Long> selectedPa
*/
@Override
public LogicalOlapTableStreamScan withSelectedPartitionIds(List<Long>
selectedPartitionIds,
- boolean
isPartitionPruned) {
+ boolean
hasPartitionPredicate) {
return AbstractPlan.copyWithSameId(this, () ->
new LogicalOlapTableStreamScan(relationId, (Table) table,
qualifier,
groupExpression, Optional.of(getLogicalProperties()),
- selectedPartitionIds, isPartitionPruned,
hasPartitionPredicate, selectedTabletIds,
+ selectedPartitionIds, true, hasPartitionPredicate,
selectedTabletIds,
Review Comment:
Although GitHub currently marks this thread resolved, rechecking current
head `72b7474d41f01a654d940c6737b51a43d6fa12f8` shows that the code issue is
still present. `NormalizeOlapTableStreamScan` still creates every base/binlog
child through a `LogicalOlapScan` constructor whose selected partition IDs are
also recorded as `specifiedPartitions`. The same top-down batch then descends
into those fresh children, treats the internal IDs as a manual partition
filter, and sets `hasPartitionPredicate=true` before physical translation—even
when the original stream query had no partition predicate.
The new test stops after `PlanChecker.analyze(...)` and invokes
`withSelectedPartitionIds(..., false)` directly, so it verifies the local
builder fix but does not exercise normalization or the final `OlapScanNode`
flag. Please preserve the already-computed partition-predicate value both when
`PruneEmptyPartition` narrows the selected IDs and when normalization
constructs executable children. Coverage should include final
filtered/unfiltered behavior plus a filtered case where only some selected
partitions are empty. I am treating this existing issue as blocking rather than
opening a duplicate inline comment.
--
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]
