[
https://issues.apache.org/jira/browse/SPARK-16733?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15395163#comment-15395163
]
Reynold Xin commented on SPARK-16733:
-------------------------------------
Putting the comment here as well:
So I think the proper fix is that in the future, we can introduce a config flag
to specifies the parse mode for column names:
1. Consistent with Spark 1.x and 2.0
2. Consistent with SQL
3. Treat as string literals.
> use antlr to parse column name instead of implementing it ourselves
> -------------------------------------------------------------------
>
> Key: SPARK-16733
> URL: https://issues.apache.org/jira/browse/SPARK-16733
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Reporter: Wenchen Fan
> Assignee: Wenchen Fan
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
