[
https://issues.apache.org/jira/browse/KAFKA-18934?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17933164#comment-17933164
]
Ismael Juma commented on KAFKA-18934:
-------------------------------------
Sounds good, thanks for picking them up. :) Generally speaking, I assign things
to myself if I intend to work on them. Always fine to pick up unassigned
tickets I create.
> Consider making records in FetchResponse not nullable
> -----------------------------------------------------
>
> Key: KAFKA-18934
> URL: https://issues.apache.org/jira/browse/KAFKA-18934
> Project: Kafka
> Issue Type: Improvement
> Reporter: Ismael Juma
> Assignee: TengYao Chi
> Priority: Major
> Labels: needs-kip
>
> We had originally done this via KIP-896, but then reverted it via
> [https://github.com/apache/kafka/pull/19131] for the reasons stated in that
> pull request.
> Jun suggested a potential way forward:
> {code:java}
> The inconsistency in the current implementation is not ideal though and it
> would be useful to improve it. One potential way is to bump up the
> FetchRequest version and make records officially non-nullable. If we do
> that, we probably want to do the same for
> FetchResponse.AbortedTransactions, ShareFetchResponse.records and
> potentially ProduceRequest.records for better consistency.{code}
> We need to decide if this is worth it.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
