mjsax commented on code in PR #11896:
URL: https://github.com/apache/kafka/pull/11896#discussion_r848932512
##########
streams/src/main/java/org/apache/kafka/streams/kstream/internals/KStreamWindowAggregate.java:
##########
@@ -184,6 +247,75 @@ public void process(final Record<KIn, VIn> record) {
droppedRecordsSensor.record();
}
}
+
+ maybeMeasureLatency(() -> tryEmitFinalResult(record, closeTime),
time, emitFinalLatencySensor);
+ }
+
+ private void tryEmitFinalResult(final Record<KIn, VIn> record, final
long closeTime) {
+ if (emitStrategy.type() != StrategyType.ON_WINDOW_CLOSE) {
+ return;
+ }
+
+ final long now = internalProcessorContext.currentSystemTimeMs();
+ // Throttle emit frequency
+ if (now < timeTracker.nextTimeToEmit) {
+ return;
+ }
+
+ // Schedule next emit time based on now to avoid the case that if
system time jumps a lot,
+ // this can be triggered everytime
+ timeTracker.nextTimeToEmit = now;
+ timeTracker.advanceNextTimeToEmit();
+
+ // Close time does not progress
+ if (lastEmitCloseTime != ConsumerRecord.NO_TIMESTAMP &&
lastEmitCloseTime >= closeTime) {
Review Comment:
Maybe. Sounds like a weird corner case that could lead to duplicate output
(what is ok for ALOS)? -- Might also be ok to just leave it in -- seems there
is no concern about correctness?
--
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]
