Github user wuchong commented on the issue:
https://github.com/apache/flink/pull/4183
Hi @fhueske , I like the `firstResultTimeOffset`, a good design ! ð
Regarding to the watermark and timestamp, it makes sense to me. And a
simple approach come to my mind : assign a new
`AssignerWithPunctuatedWatermarks` after window aggregate. This will create an
operator which ignores the upstream watermarks and assign new watermarks depend
on the watermark function.
The new watermark function can simply take the element's timestamp as the
watermark.
```scala
class TimestampAndWatermark[T] extends AssignerWithPunctuatedWatermarks[T] {
override def checkAndGetNextWatermark(lastElement: T,
extractedTimestamp: Long): Watermark = {
new Watermark(extractedTimestamp)
}
override def extractTimestamp(element: T, originalTimestamp: Long):
Long = originalTimestamp
}
```
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
