aljoscha commented on issue #8535: [FLINK-11693] Add KafkaSerializationSchema 
that uses ProducerRecord
URL: https://github.com/apache/flink/pull/8535#issuecomment-505488291
 
 
   I updated the PR, PTAL (please take another look 😅)
   
   Regarding partitioning: yes, this is a shortcoming in the current design and 
I didn't want to tackle it because I see the same two options that you 
mentioned and both require some work. In general I think the fact that we have 
our own serializers and always use `ByteArrayDeserializer` (and serializers) 
for Kafka is a bit weird and means that people can't use the same serializers 
that they use elsewhere with Kafka.
   
   Regarding your option 1., I think this is tricky, because the partition that 
we partitioner returns has to be put into the Kafka `ProducerRecord`. This 
worked before because the framework code was assembling the `ProducerRecord`. 
When the `KafkaSerializationSchema` returns a `ProducerRecord`, this is not 
possible because partition (and the other fields) are `final`.

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to