bharatviswa504 commented on issue #825: URL: https://github.com/apache/hadoop-ozone/pull/825#issuecomment-617344891
> Thanks to start this discussion @maobaolong. I agree with @bharatviswa504 , I like when the de-serialization and the serialization logic is close to the data object itself and I don't like the big utility classes. > > But I also understand your arguments (except the first, one, I think if the code is read in Intellij, it's a hard situation anyway). > > What do you think about introducing a companion implementation / interface similar to `org.apache.hadoop.hdds.utils.db.Codec`. With this approach we can keep the serialization logic close to the data objects but they can be in a separated class. I like the proposal of the interface, and move the logic of protobuf serialization to that class. In this way, it is closer to objects. The main reason for my comment is the same as @elek if we have the logic of protobuf in a single class, it will become one huge class and to follow it will be hard. ---------------------------------------------------------------- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
