See https://beam.apache.org/documentation/programming-guide/#inferring-schemas for details on how to use schemas with various types of Java classes (POJO, Java Beans, AutoValue).
On Mon, Mar 22, 2021 at 3:01 PM Robert Bradshaw <rober...@google.com> wrote: > Thanks, that looks quite useful. > > A not on POJOs, Serializable has several disadvantages (e.g. > non-deterministic coding for key grouping, less efficient serialization). > You could look into making them compatible with Beam schemas. > > On Mon, Mar 22, 2021 at 1:11 PM Cristian Constantinescu <zei...@gmail.com> > wrote: > >> Hi everyone, >> >> I spent the week-end putting the pieces together to run Beam with the >> Flink Runner on Kubernetes. While I did find very good articles and videos >> about Beam and Flink and Kubernetes separately, I didn't find one that >> mixes all three of them in the same pot. >> >> So, I wrote a small demo project/tutorial combining all the bits and >> pieces of information I found on my quest. It's here >> https://github.com/zeidoo/tutorial-bean-flink-kubernetes-pojo. Any >> feedback is welcome. >> >> I hope it helps someone. >> >> Cheers everyone! >> >> PS: I didn't find on the Beam documentation website that the easiest way >> to pass PoJos around transforms is to make them implement Serializable. >> I've put a small section about that in the tutorial. >> >
