This would be Michael and Guozhang job to answer, but I'd look at two options if I were you:
1) If the connector you need exists ( http://www.confluent.io/product/connectors), then you just need to run it. It is just a simple REST API (submit job, job status, etc), so I wouldn't count it as "learning a framework". 2) I believe I've seen people implement "writes to database" in KafkaProcessor. Maybe try to google / search the mailing list? Guozhang and Michael can probably add details. Gwen On Wed, Jun 1, 2016 at 11:31 PM, Avi Flax <avi.f...@parkassist.com> wrote: > On 6/1/16, 11:53, "Gwen Shapira" <g...@confluent.io> wrote: > > > Currently this is not part of the DSL and needs to be done separately > > through KafkaConnect. Here's an example: > > http://www.confluent.io/blog/hello-world-kafka-connect-kafka-streams > > Ah, I see! Thank you! And thanks for the super-fast reply! > > > In the future we want to integrate Connect and Streams better, so you > could > > do something like "builder.stream(..).fromCassandra()" and > > "stream.toElastic()" and a connector will be added that does the "last > > mile" of data integration. > > That sounds great! > > In the meantime, while I understand that Connect is the recommended > approach for this sort of thing, I’d like to avoid learning two frameworks > simultaneously if possible — and not just for me, but also for my team; is > it possible to use the low-level Streams API to implement the sources and > sinks I’ve described? And if so, is it possible to integrate sources and > sinks defined with that lower-level API with a topology defined with the > DSL? > > Thanks so much! > Avi > > Software Architect @ Park Assist » We’re hiring! > >
