Github user shaoxuan-wang commented on the issue:
https://github.com/apache/flink/pull/3470
@fhueske, we should add different built-in aggregates for the ones that are
not easy to be retracted (for instance Max and Min), one as withRetract(used
for over window as # of items are limited, also for the dataStream when
retraction is absolute needed) the other one as withoutRetract (used for
dataSet and dataStream where retraction is not needed), it is the optimizer to
check the rules and decide which aggregate should be used. I think we can go
ahead as this (withRetract) for now, and complete the entire design together
with dataStream retraction.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
