[
https://issues.apache.org/jira/browse/SPARK-20202?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16500560#comment-16500560
]
Steve Loughran commented on SPARK-20202:
----------------------------------------
I think you could split things into two
# a modified hive 1.2.1.x for hadoop 3, with a new package name in the maven
builds. (joy, profiles!) and some work with the hive team to get this
officially published by them. Strength: easy for people to backport into
shipping 2.2, 2.3 builds just by changing the POM
# the bigger move to Hive 2. This will be the best for future, but is bound to
have more surprises. There's even the possibility that the hive team might have
to make some changes too, which isn't impossible if the timelines line up.
> Remove references to org.spark-project.hive
> -------------------------------------------
>
> Key: SPARK-20202
> URL: https://issues.apache.org/jira/browse/SPARK-20202
> Project: Spark
> Issue Type: Bug
> Components: Build, SQL
> Affects Versions: 1.6.4, 2.0.3, 2.1.1
> Reporter: Owen O'Malley
> Priority: Major
>
> Spark can't continue to depend on their fork of Hive and must move to
> standard Hive versions.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
