Github user massie commented on a diff in the pull request:
https://github.com/apache/spark/pull/7403#discussion_r38469436
--- Diff: core/src/main/scala/org/apache/spark/rdd/PairRDDFunctions.scala
---
@@ -565,12 +603,25 @@ class PairRDDFunctions[K, V](self: RDD[(K, V)])
}
/**
- * Simplified version of combineByKey that hash-partitions the resulting
RDD using the
- * existing partitioner/parallelism level.
+ * This method is here for backward compatibility. It
+ * does not provide combiner classtag information to
+ * the shuffle.
+ *
+ * @see [[combineByKeyWithClassTag]]
*/
def combineByKey[C](createCombiner: V => C, mergeValue: (C, V) => C,
mergeCombiners: (C, C) => C)
- : RDD[(K, C)] = self.withScope {
- combineByKey(createCombiner, mergeValue, mergeCombiners,
defaultPartitioner(self))
+ : RDD[(K, C)] = {
+ combineByKeyWithClassTag(createCombiner, mergeValue,
mergeCombiners)(null)
+ }
+
+ /**
+ * Simplified version of combineByKeyWithClassTag that hash-partitions
the resulting RDD using the
+ * existing partitioner/parallelism level.
+ */
+ def combineByKeyWithClassTag[C](createCombiner: V => C, mergeValue: (C,
V) => C,
+ mergeCombiners: (C, C) => C)
+ (implicit ct: ClassTag[C]): RDD[(K, C)] = self.withScope {
--- End diff --
The original `combineByKey` methods had `self.withScope` so the new
`combineByKeyWithClassTags` should too, I believe.
I was worried about nesting the scopes but I think that may be beneficial
(and I'll take your advice to re-add them). We want users to be able use the UI
to debug if their application is using the `combineByKey` or the
`combineByKeyWithClassTags` methods, since some shuffle implementation will
need `ClassTag` information.
---
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]
