Github user fhueske commented on the issue:
https://github.com/apache/flink/pull/3641
You are right, with processing time no two records arrive at exactly the
same time. However, we also retract values based on the granularity of
milliseconds. Therefore, I think is makes sense to also do the accumulation on
this granularity. The other reason for the change would be consistency with
event-time as you mentioned.
The `ProcessFunction` does also support to register processing time timers
via the context object.
With the harness we decouple the processing time from the system time. I
would hope that the harness also triggers a timer when we increment the
processing time but I'm not 100% sure it does.
---
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.
---
