Hi Frank, As far as I know the design in that wiki has been superceded by the Global KTables design which is now coming in 0.10.2. Hence, the JIRAs that are mentioned there (like KAFKA-3705). There are some extensive comments in https://issues.apache.org/jira/browse/KAFKA-3705 <https://issues.apache.org/jira/browse/KAFKA-3705> illustrating why this design is particularly challenging and why Global KTables was chosen instead. I'm not sure if you still want to pursue that original design, since it is not proven to work.
Guozhang, perhaps we need to add a note saying that Global KTables is the new design? Thanks Eno > On 21 Feb 2017, at 07:35, Frank Lyaruu <flya...@gmail.com> wrote: > > Hi all, > > I'm trying to implement joining two Kafka tables using a 'remote' key, > basically as described here: > > https://cwiki.apache.org/confluence/display/KAFKA/Discussion%3A+Non-key+KTable-KTable+Joins > > Under the "Implementation Details" there is one line I don't know how to > do: > > > 1. First of all, we will repartition this KTable's stream, by key > computed from the *mapper(K, V) → K1*, so that it is co-partitioned by > the same key. The co-partition topic is partitioned on the new key, but the > message key and value are unchanged, and log compaction is turned off. > > > How do I do that? I've been unable to find any documentation, I've looked > at the StreamPartitionAssignor, that seems relevant, but I could use some > help. Does anyone have an example? > > regards, Frank
