[
https://issues.apache.org/jira/browse/KAFKA-8288?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16826166#comment-16826166
]
Matthias J. Sax commented on KAFKA-8288:
----------------------------------------
This behavior is by design. Compare:
https://issues.apache.org/jira/browse/KAFKA-4785
I agree that the docs are a little bit miss-leading. Do you face a particular
issue with the behavior? Or are you just surprised about it, because the docs
are not 100% accurate?
> KStream.through consumer does not use custom TimestampExtractor
> ---------------------------------------------------------------
>
> Key: KAFKA-8288
> URL: https://issues.apache.org/jira/browse/KAFKA-8288
> Project: Kafka
> Issue Type: Bug
> Components: streams
> Affects Versions: 2.2.0
> Reporter: Maarten
> Priority: Minor
>
> The Kafka consumer created by {{KStream.through}} does not seem to be using
> the custom TimestampExtractor set in Kafka Streams properties.
> The documentation of {{through}} states the following
> {code:java}
> ...
> This is equivalent to calling to(someTopic, Produced.with(keySerde,
> valueSerde) and StreamsBuilder#stream(someTopicName, Consumed.with(keySerde,
> valueSerde)).
> {code}
> However when I use the pattern above, the custom TimestampExtractor _is_
> called.
> I have verified that the streams app is reading from the specified topic and
> that the timestamp extractor is called for other topics.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
