[
https://issues.apache.org/jira/browse/SPARK-17053?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15420507#comment-15420507
]
Reynold Xin edited comment on SPARK-17053 at 8/14/16 10:32 PM:
---------------------------------------------------------------
It was by accident that this was supported, because Spark simply used Hive's
code to do all DDL operations. In Spark 2.0, Spark implemented all the DDLs
natively in Spark, and it is not the intention of the project to support all
the esoteric feature flags provided by Hive.
was (Author: rxin):
It was by accident that this was supported, because Spark simply used Hive's
code to do all DDL operations. In Spark 2.0, Spark implemented all the DDLs
natively in Spark, and it is not the intention of the project to support all
the esoteric features provided by Hive.
> Spark ignores hive.exec.drop.ignorenonexistent=true option
> ----------------------------------------------------------
>
> Key: SPARK-17053
> URL: https://issues.apache.org/jira/browse/SPARK-17053
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 2.0.0
> Reporter: Gokhan Civan
>
> In version 1.6.1, the following does not throw an exception:
> create table a as select 1; drop table a; drop table a;
> In version 2.0.0, the second drop fails; this is not compatible with Hive.
> The same problem exists for views.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
