jerryshao commented on PR #11997: URL: https://github.com/apache/gravitino/pull/11997#issuecomment-4989004460
Hi @krickert thanks a lof for your work. Yes, we left the room to support schema in the future, but we haven't implemented it till now. The reason why we haven't done this part is that we still have several usage scenario questions: 1. Users may already have a schema registry. How to work with the existing schema registry? Which one is the SSOT? How do we keep feature parity if we want to support this? 2. How do we expose this feature, which means how do we expose the APIs for producer/consumer to use, even further, how do we leverage streaming engines like Flink? These leave room for us to discuss before implementation. Anyway, I think you did a very good start. What I suggest are the next steps: 1. Converting this PR/issue into a design doc. Bring this out to the dev mail list to involve others to review and discuss. 2. Once the design is finalized, you can split it into multiple subtasks and submit PR one by one. Typically, it is hard to review a very big PR, and we request contributors to split it into smaller ones to accelerate the process. So, overall, a typical way to contribute a new feature is to discuss the design first. Thanks. Also, cc @mchades to take a look at this PR. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
