Also, there is a couple of things that should be noted. @vvcephei In your code, it appears that you added a ```Map``` to allow us to tell which ```TaskId``` matches with what ```TaskMetadata```. But such an approach could be difficult to maintain because ```TaskId``` occurs several times as fields throughout the code. Therefore, each time ```TaskId``` occurs as a field, some change to the current fields or adding a new field would be required to accomadate for that. In contrast, inheritance allows us to naturally embed the metadata into the existing fields. Inheritance would allow further upgrades in the future easier, meanwhile, the scattered manner in which you keep tabs on ```TaskMetadata``` would be harder to keep track of (your way might also have heavier memory consumption i.e. you are adding a new Map field to keep track of the data).
[ Full content available at: https://github.com/apache/kafka/pull/5012 ] This message was relayed via gitbox.apache.org for [email protected]
