[
https://issues.apache.org/jira/browse/SPARK-10550?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14740959#comment-14740959
]
shao lo commented on SPARK-10550:
---------------------------------
I do not know if this is valid Scala, but something like...
@transient private[sql] lazy val listener = new SQLListener(this){
sparkContext.addSparkListener(listener)
sparkContext.ui.foreach(new SQLTab(this, _))
}
What I really need is a way to initialize an extended SQLContext with a
reference to an already initialized SQLContext. In 1.4.1 I was able to do the
following..
public class MySQLContext extends SQLContext
{
SparkContext sc;
public MySQLContext(JavaSparkContext sc)
{
super(sc.sc());
this.sc = sc.sc();
}
}
SparkContext ctx1 = new SparkContext(sparkConf);
JavaSparkContext ctx = new JavaSparkContext(ctx1);
MySQLContext sqlCtx = new MySQLContext(ctx);
> SQLListener error constructing extended SQLContext
> ---------------------------------------------------
>
> Key: SPARK-10550
> URL: https://issues.apache.org/jira/browse/SPARK-10550
> Project: Spark
> Issue Type: Bug
> Components: SQL
> Affects Versions: 1.5.0
> Reporter: shao lo
> Priority: Minor
>
> With spark 1.4.1 I was able to created a custom SQLContext class. With spark
> 1.5.0, I now get an error calling the super class constructor. The problem
> is related to this new code that was added between 1.4.1 and 1.5.0
> // `listener` should be only used in the driver
> @transient private[sql] val listener = new SQLListener(this)
> sparkContext.addSparkListener(listener)
> sparkContext.ui.foreach(new SQLTab(this, _))
> ..which generates
> Exception in thread "main" java.lang.NullPointerException
> at
> org.apache.spark.sql.execution.ui.SQLListener.<init>(SQLListener.scala:34)
> at org.apache.spark.sql.SQLContext.<init>(SQLContext.scala:77)
--
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
