Github user fanyon commented on the issue:
https://github.com/apache/flink/pull/3766
@gyfora For the method `int partition(T next, byte[] serializedKey, byte[]
serializedValue, int partitionNum)` in KafkaPartitioner, the correct partition
num of target topic can be used. But the KafkaPartitioner's partition id array
has been initialized in `void open(int parallelInstanceId, int
parallelInstances, int[] partitions)`, which will be executed once, so yes, the
problem for dynamic new topics when user uses older KafkaPartitioner API in
their older job will still exist, and I find it hard to solve this problem
completely.
What do you think of this? @tzulitai
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
