Github user rmetzger commented on the pull request:
https://github.com/apache/incubator-flink/pull/243#issuecomment-65075505
Thank you for reviewing the pull request!
I would suggest to first merge the change and then start improving it by
changing the internal types.
Regarding the 2. point (POJO support). I have once started working on that.
I think the best approach would be to add a method to the serializers which
allows to access fields.
By doing so, we can re-use the existing code for analyzing (nested) POJO
fields. Also, it would generalize between Tuples and POJOs and avoid special
casing between Tuples and POJOs (it is trivial to implement field access for
Tuples in the Tuple serializers)
Once the serialisation for POJOs has become more efficient, we can actually
throw away the Tuple specific code and handle them like POJOs.
---
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.
---
