Github user JoshRosen commented on a diff in the pull request:
https://github.com/apache/spark/pull/3711#discussion_r22231341
--- Diff:
core/src/main/scala/org/apache/spark/scheduler/SparkListener.scala ---
@@ -84,6 +86,14 @@ case class SparkListenerBlockManagerRemoved(time: Long,
blockManagerId: BlockMan
@DeveloperApi
case class SparkListenerUnpersistRDD(rddId: Int) extends SparkListenerEvent
+@DeveloperApi
+case class SparkListenerExecutorAdded(executorId: String, executorInfo :
ExecutorInfo)
--- End diff --
I'm on the fence about whether this should be a case class or a regular
class. If it's a case class, then Scala users can pattern-match on it, but we
risk introducing binary incompatibilities if we ever need to add new fields to
this event. If we had to re-do the SparkListener API, I'd probably opt to not
use case classes for this reason. In light of this, do you think we should we
adopt a "no new cases classes" policy for new listener events?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
