Ok, I've decided that this optimization is premature and complex, so I've gone ahead and simplified it. (I'll let you know when I push the update).
I've also updated the TimeDefinition class to be less flexible (although it doesn't really simplify this particular method). FWIW, though I think that "suppress for 0ms" is a perfectly valid way to disable a suppression operation. Note that this is also what we wind up with when you use final-results on a windowed stream with gracePeriod set to 0ms, which also seems perfectly fine. [ Full content available at: https://github.com/apache/kafka/pull/5693 ] This message was relayed via gitbox.apache.org for [email protected]
