cadonna commented on code in PR #13855:
URL: https://github.com/apache/kafka/pull/13855#discussion_r1235096344
##########
streams/src/main/java/org/apache/kafka/streams/kstream/internals/KStreamImpl.java:
##########
@@ -1256,10 +1261,25 @@ private <VO, VR> KStream<K, VR> doStreamTableJoin(final
KTable<K, VO> table,
final NamedInternal renamed = new NamedInternal(joinedInternal.name());
final String name = renamed.orElseGenerateWithPrefix(builder, leftJoin
? LEFTJOIN_NAME : JOIN_NAME);
+
+ Optional<TimeOrderedKeyValueBuffer<K, V, V>> buffer = Optional.empty();
+
+ if (joined.gracePeriod() != null) {
Review Comment:
> So if someone set the grace period to zero then we can evict everything
from the buffer even after the grace period is changed.
Are you saying that if I set the grace period > 0 and I run my application,
then I stop my application and set the grace period == 0, and restart my
application, my application will process records from my first run?
Why should removing the grace period (grace period == null) behave
differently?
Actually we do not support topology updates. However, we can at least make
equivalent changes behave equivalently.
--
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]
