You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
One problem I encountered was when I tried to test a TumblingWindow aggregation.
Let's say I create a 2 second tumbling window like so: val w = TimeWindows .of(TimeUnit.SECONDS.toMillis(2)) .advanceBy(TimeUnit.SECONDS.toMillis(2)) .until(TimeUnit.SECONDS.toMillis(2))
Now, I use the CustomTimestampExtractor and send in events with value "number@timestamp", e.g.
("a", "1@1000"), ("a", "1@1500"), ("a", "1@2000"), ("a", "1@10000"), ("a", "1@100")
I run a simple count aggregation and would expect count to be (0->2), (2000->1), (10000->1), because the last event arrives when the time window [0,2000) is already closed because of until(2000). However, the last event is accounted for.
Could this be an interpretation, that the until duration is understood as a lower bound?
Thanks
The text was updated successfully, but these errors were encountered:
Hi, thanks for the great library!
One problem I encountered was when I tried to test a TumblingWindow aggregation.
Let's say I create a 2 second tumbling window like so:
val w = TimeWindows .of(TimeUnit.SECONDS.toMillis(2)) .advanceBy(TimeUnit.SECONDS.toMillis(2)) .until(TimeUnit.SECONDS.toMillis(2))
Now, I use the CustomTimestampExtractor and send in events with value "number@timestamp", e.g.
("a", "1@1000"), ("a", "1@1500"), ("a", "1@2000"), ("a", "1@10000"), ("a", "1@100")
I run a simple count aggregation and would expect count to be (0->2), (2000->1), (10000->1), because the last event arrives when the time window [0,2000) is already closed because of until(2000). However, the last event is accounted for.
Could this be an interpretation, that the until duration is understood as a lower bound?
Thanks
The text was updated successfully, but these errors were encountered: