Github user fhueske commented on the issue:
https://github.com/apache/flink/pull/6201
Hi @suez1224, that sounds good overall. :-)
A few comments:
- I would not add a user-facing property `connector.support-timestamp`
because a user chooses that by choosing the connector type. Whether the
connector supports writing a system timestamp can be an internal
field/annotation/interface of the `TableSink` that is generated from the
properties.
- Copying the timestamp to the StreamRecord timestamp field can be done
with a process function. Actually, we do that already when converting a Table
into a DataStream. Setting the flag in the Kafka TableSink should be easy.
- Not sure if `from-source` needs to be supported by the initial version.
We could just implement `from-field` for now, and handle `from-source` as a
follow up issue. Since we are approaching feature freeze, I think this might be
a good idea at this point.
What do you think?
Fabian
---
