Cyrill commented on code in PR #2679:
URL: https://github.com/apache/ignite-3/pull/2679#discussion_r1362007626
##########
modules/table/src/main/java/org/apache/ignite/internal/table/distributed/replicator/PartitionReplicaListener.java:
##########
@@ -1763,12 +1768,15 @@ private CompletableFuture<ReplicaResult>
processMultiEntryAction(ReadWriteMultiR
return allOf(deleteExactLockFuts).thenCompose(ignore -> {
Map<UUID, BinaryRowMessage> rowIdsToDelete = new
HashMap<>();
Collection<BinaryRow> result = new ArrayList<>();
+ List<RowId> rows = new ArrayList<>();
for (int i = 0; i < searchRows.size(); i++) {
RowId lockedRowId = deleteExactLockFuts[i].join();
if (lockedRowId != null) {
rowIdsToDelete.put(lockedRowId.uuid(), null);
+
+ rows.add(lockedRowId);
Review Comment:
This logic tracks down all existing rows that a transaction may affect.
In the case of INSERT_ALL we have
```
rowsToInsert.put(new RowId(partId(), UUID.randomUUID()), row);
```
which means we create completely new rows (obviously it is what an insert
does) - thus no need to cleanup them since by every chance these rows will not
be present in the storage.
--
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]
