Debasish Ghosh

Please submit a PR to apache/kafka. `streams-scala` is on `master` now .. https://github.com/apache/kafka/tree/trunk/streams/streams-scala .. and will be released with 2.0

Why do we need this ? We don't want to create any unnecessary wrappers. The Java version works nicely with Scala ..

We plan to gradually improve upon the coverage of the Scala APIs and state store APIs are definitely an option. But at the moment we don't have any committed timeline...

The APIs that take `Consumed` can always pass an instance of `Consumed` with `TimestampExtractor`. I am not sure I understand what the wrapper API needs to provide here.

I suggest you ask this question on kafka user mailing list. This Scala wrapper does nothing more than wrapt the Java APIs - hence it is unlikely someone will be...

Can u please clarify what u mean by *use kafka-streams-scala with case classes* ? Maybe an example of what u want to do ..

Currently we don't have higher level APIs towards this end.

The example with `User` case class applies to the current library as well. You need to define the `Serde` for the case class and pass it along. I was talking...

Hi @mazorigal - In the latest release `0.2.0` we have implicit serdes implementation, where u can define a case class, define its serde, have an implicit in scope and then...

We kept it `AT_LEAST_ONCE` to keep the delivery semantics same as the other runtimes, Akka and Spark. May be we can change it since Flink supports `EXACTLY_ONCE` - WDYT @RayRoestenburg...