github-actions[bot] commented on code in PR #66913:
URL: https://github.com/apache/doris/pull/66913#discussion_r3870564423
##########
fe/fe-core/src/main/java/org/apache/doris/datasource/hudi/source/HudiScanNode.java:
##########
@@ -521,23 +591,29 @@ private void getPartitionsSplits(List<HivePartition>
partitions, List<Split> spl
@Override
public List<Split> getSplits(int numBackends) throws UserException {
- if (incrementalRead && !incrementalRelation.fallbackFullTableScan()) {
- return getIncrementalSplits();
- }
- initPrunedPartitions();
- List<Split> splits = Collections.synchronizedList(new ArrayList<>());
+ ensureHmsRuntimeGeneration();
+ acquireFsView();
Review Comment:
[P1] Skip the unused fs-view sync for incremental reads
This branch now runs only after `acquireFsView()`, whose `tryAcquire()`
synchronously calls `fsView.sync()`. None of the non-fallback incremental
implementations uses `HudiScanNode.fsView`: COW consumes its resolved path
sets, MOR constructs a separate `HoodieTableFileSystemView`, and the empty
relation returns immediately. Consequently an incremental query can now block
or fail on an unrelated shared-view refresh that its split path never reads.
Take this branch before acquiring the cached view (while retaining the HMS
generation checks), and cover COW, MOR, and empty incremental reads.
--
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]
