Github user cloud-fan commented on a diff in the pull request:
https://github.com/apache/spark/pull/16977#discussion_r102357800
--- Diff:
core/src/main/scala/org/apache/spark/rdd/ParallelCollectionRDD.scala ---
@@ -105,6 +105,17 @@ private[spark] class ParallelCollectionRDD[T:
ClassTag](
override def getPreferredLocations(s: Partition): Seq[String] = {
locationPrefs.getOrElse(s.index, Nil)
}
+
+ override def collect(): Array[T] = toArray(data)
+
+ override def take(num: Int): Array[T] = toArray(data.take(num))
+
+ private def toArray(data: Seq[T]): Array[T] = {
+ // We serialize the data and deserialize it back, to simulate the
behavior of sending it to
+ // remote executors and collect it back.
+ val ser = sc.env.closureSerializer.newInstance()
+ ser.deserialize[Seq[T]](ser.serialize(data)).toArray
+ }
--- End diff --
to strictly follow the previous behavior: users may still have the
reference to the `data` and change it, but the result of `collect` should not
be affected.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
