[
https://issues.apache.org/jira/browse/SPARK-9292?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Reynold Xin updated SPARK-9292:
-------------------------------
Issue Type: Sub-task (was: Bug)
Parent: SPARK-8947
> Analysis should check that join conditions' data types are booleans
> -------------------------------------------------------------------
>
> Key: SPARK-9292
> URL: https://issues.apache.org/jira/browse/SPARK-9292
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Reporter: Josh Rosen
> Assignee: Josh Rosen
>
> The following data frame query should fail analysis but instead fails at
> runtime:
> {code}
> val df = Seq((1, 1)).toDF("a", "b")
> df.join(df, df.col("a"))
> {code}
> This should fail with an AnalysisException because the column "A" is not a
> boolean and thus cannot be used as a join condition.
> This can be fixed by adding a new analysis rule which checks that the join
> condition has BooleanType.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
