It's great. Thank you so much for leading this, Panos.
+1 for starting Apache Spark 1.6.7 release.
I'm also expecting a release for the following patch which was merged last
week.
ORC-711: Support CryptoExtension in create/decryptLocalKey
For now, the current snapshot seems to keep working well with Apache Spark
and Apache Iceberg, too.
I'll volunteer for a release manager for 1.6.7 and cut a release next
Monday if there are no other issues.
Bests,
Dongjoon.
On Fri, Jan 15, 2021 at 3:08 PM Panos Garefalakis <[email protected]>
wrote:
> Hello all,
>
> Keeping up the momentum from the recent backwards compatible Spark release
> (1.6.6), I would like to propose a new 1.6.7 release as the fist compatible
> version for Apache Hive!
> To achieve this, the community had to resolve a number of issues including:
>
> * ORC-705 <https://issues.apache.org/jira/browse/ORC-705>: Predicate
> evaluation taking into account writer calendar
> * ORC-706 <https://issues.apache.org/jira/browse/ORC-706>: Putting back
> DataReaderProperties default maxDiskRangeChunkLimit
> * ORC-709 <https://issues.apache.org/jira/browse/ORC-709>: Boolean to
> StringGroup schema evolution
> * ORC-724 <https://issues.apache.org/jira/browse/ORC-724>: PPD: IN single
> value comparison throwing ClassCastException with Date objects
>
> As of yesterday Hive precommit tests are also passing using the latest
> 1.6.7-SNAPSHOT version and Hive community is already discussing the upgrade
> -- check HIVE-23553 <https://issues.apache.org/jira/browse/HIVE-23553> for
> the implementation details.
>
> Is there anything else that should also go in?
> Please let me know what you think!
>
> Cheers,
> Panagiotis
>
