Hi,
Is joining 2 streams based on Windowed<String> keys supposed to work?
I have 2 KTables:
- KTable<Windowed<String>, T> events: I process events and aggregate events
that have a common criteria using aggregateByKey and UnlimitedWindows as window
(for now)
- KTable<Windowed<String>, S> hourlyStats: I calculate some stats using
aggregateByKey for hourly windows TimeWindows.of(“window name”, hourly)
Since both use aggregateByKey() they are both KTables and both have
Windowed<String> keys.
I need to leftJoin the first one (events) with the second one (hourlyStats) BUT
I need to join event x which occurred at time t0 with hourlyStats of the t0
window. In other words I need to join using
JoinWindows.of("JoinWindow").within(60 * 60 * 1000). Since both are KTables
this is not possible. But if I turn them both into KStream's using toStream()
then I can use the leftJoin() variants which supports JoinWindows. They would
both be KStream<Windowed<String>, ...>.
The problem is the join doesn’t really happen. No hourlyStats is actually found
for any row of events. The TimestampExtractor for both is correct.
So, Is joining 2 streams based on Windowed<String> keys supposed to work? If
not then how can I accomplish the above task?
Thanks,
Ara.
________________________________
This message is for the designated recipient only and may contain privileged,
proprietary, or otherwise confidential information. If you have received it in
error, please notify the sender immediately and delete the original. Any other
use of the e-mail by you is prohibited. Thank you in advance for your
cooperation.
________________________________