Github user chenghao-intel commented on the pull request:
https://github.com/apache/spark/pull/7520#issuecomment-135245692
Actually I have some concerns on the change.
1. Using the StructObjectInspector is harmless for serialization, exactly
the same as using the OrcStruct, and associated with an
`OrcStructInspector`.You can check the source code at
`org.apache.hadoop.hive.ql.io.orc.WriterImpl`, it's still requires the
`StructObjectInspector` for writing rows. And it will be great if you can
provide the performance improvement says OrcStruct is much more efficient than
StructObjectInspector in serialization.
2. Hive is actually a case-insensitive system, normalizing the name(to
lower case) seems the default behaviour of Hive, since we are supporting(#5733)
seamless integrated ORC with Hive, we'd better follow the same critiera,
otherwise probably cause some confusing problem.
Sorry if I missed some of your original motive of the change.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
