[GitHub] [kafka] vvcephei commented on a diff in pull request #12803: KAFKA-13602: Adding ability to multicast records.

[GitBox <gi...@apache.org>]

vvcephei commented on code in PR #12803:
URL: https://github.com/apache/kafka/pull/12803#discussion_r1037554572


##########
streams/src/main/java/org/apache/kafka/streams/processor/internals/StreamsMetadataState.java:
##########
@@ -459,12 +461,23 @@ private List<StreamsMetadata> rebuildMetadataForSingleTopology(final Map<HostInf
         return rebuiltMetadata;
     }
 
+    private final Function<Optional<Set<Integer>>, Integer> getPartition = maybeMulticastPartitions -> {
+        if (!maybeMulticastPartitions.isPresent()) {
+            return null;
+        }
+        if (maybeMulticastPartitions.get().size() != 1) {
+            throw new IllegalArgumentException("The partitions returned by StreamPartitioner#partitions method when used for fetching KeyQueryMetadata for key should be a singleton set");

Review Comment:
   Hi all, on question (2), I'm not sure that it'll be straightforward to establish the semantics of FK Join on multicasted tables. Personally, I'd be more comfortable just keeping the restriction that we see in this PR and following up with a separate KIP/PR to lift that restriction.
   
   I don't think that restriction will directly harm the benefit we're trying to achieve with this KIP, and it'll be relatively complicated to establish what a multicasted FK join means and whether it is correct.



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: jira-unsubscribe@kafka.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org