You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Nick Dimiduk (JIRA)" <ji...@apache.org> on 2019/04/29 18:44:00 UTC
[jira] [Commented] (SPARK-27597) RuntimeConfig should be
serializable
[ https://issues.apache.org/jira/browse/SPARK-27597?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16829612#comment-16829612 ]
Nick Dimiduk commented on SPARK-27597:
--------------------------------------
It would nice nice if there was an API built into the {{FunctionalInterface}} that let their implementation access the fully populated {{SparkSession}}.
> RuntimeConfig should be serializable
> ------------------------------------
>
> Key: SPARK-27597
> URL: https://issues.apache.org/jira/browse/SPARK-27597
> Project: Spark
> Issue Type: Improvement
> Components: SQL
> Affects Versions: 2.4.1
> Reporter: Nick Dimiduk
> Priority: Major
>
> When implementing a UDF or similar, it's quite surprising to see that the {{SparkSession}} is {{Serializable}} but {{RuntimeConf}} is not. When modeling UDFs in an object-oriented way, this leads to quite a surprise, an ugly NPE from the {{call}} site.
> {noformat}
> Caused by: java.lang.NullPointerException
> at org.apache.spark.sql.SparkSession.sessionState$lzycompute(SparkSession.scala:143)
> at org.apache.spark.sql.SparkSession.sessionState(SparkSession.scala:141)
> at org.apache.spark.sql.SparkSession.conf$lzycompute(SparkSession.scala:170)
> at org.apache.spark.sql.SparkSession.conf(SparkSession.scala:170)
> ...{noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org