Hi

Please suggest me on what is the best way to write a Sink in Beam.  I see
that there is a Sink<T> abstract class which is in experimental state.
What is the expected outcome of this one? Do we have the api frozen, or
this could still change?  Most of the existing Sink implementations like
KafkaIO.Write are not using this interface, and instead extends
PTransform<PCollection<KV<K, V>>, PDone>. Would these be changed to extend
Sink<>.


My immediate requirement is to run this Sink on FlinkRunner. Which mandates
that my implementation must also implement SinkFunction<>.  In that case,
none of the Sink<> methods get called anyway.

Regards
Sumit Chawla

Reply via email to