Github user tillrohrmann commented on the pull request:
https://github.com/apache/flink/pull/1465#issuecomment-165474222
@aljoscha is absolutely right about the performance aspect. However, the
key difference is that he was referring in his comment to the implementation of
the `serialize` method which is extensively used and, thus, should be as
efficient as possible.
If I'm not mistaken, then this argument does not apply to the
`createAuxiliaryFields` method, though. The reason is that this method is
called before the job is actually executed and, consequently, should not have
any impact on the runtime behaviour. But I might be wrong here, since we're
also shipping `TypeInformation` to the cluster, nowadays.
---
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.
---
