[
https://issues.apache.org/jira/browse/CAMEL-16060?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17317747#comment-17317747
]
Jens Kordowski commented on CAMEL-16060:
----------------------------------------
the kafka key also transports metadata information with the record. the key is
usually used to influence the partition, but not necessarily. I can see the
need for two headers to define each independently (if required). If somebody
wants to send a key-value pair to partition 0 (independent of the key), why
not, it might be required for some reason.
> camel-kafka - decouple kafka.PARTITION_KEY from kafka.KEY
> ---------------------------------------------------------
>
> Key: CAMEL-16060
> URL: https://issues.apache.org/jira/browse/CAMEL-16060
> Project: Camel
> Issue Type: Improvement
> Components: camel-kafka
> Affects Versions: 3.7.0, 2.25.3
> Reporter: Jens Kordowski
> Priority: Major
> Fix For: 3.10.0
>
>
> Currently, the camel-kafka producer requires the definition of
> kafka.PARTITION_KEY *and* kafka.KEY to specify the partition to send to.
> While the key is usually used to determine the partition to send to, I don't
> understand the hard limitation on this header, when specifically defining the
> exact partition to send to anyway (via kafka.PARTITION_KEY).
> Also, this is not clearly stated in the documentation.
> Not sure, if this qualifies as bug or improvement, looks like a limitation to
> me. I can create a PR, if desired.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
