[
https://issues.apache.org/jira/browse/SPARK-28854?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16914981#comment-16914981
]
Hao Yang Ang commented on SPARK-28854:
--------------------------------------
[~hyukjin.kwon] if xs is "consumed" at the first map then wouldn't
sc.parallelize(Seq(1, 2, 3)).mapPartitions(xs =>
xs.map(2*).map(2*)).collect.foreach(println)
fail the same way as well?
And why would zipping with empty iterators fail? The error feels more like
Iterator#hasNext was wrong about Iterator#next.
What you have suggested is just another workaround. The fact is still a bug
until zip is fixed or if you can stop users from writing the offending code at
compile-time, eg having a new iterator type that doesn't do zip.
> Zipping iterators in mapPartitions will fail
> --------------------------------------------
>
> Key: SPARK-28854
> URL: https://issues.apache.org/jira/browse/SPARK-28854
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 2.4.3
> Reporter: Hao Yang Ang
> Priority: Minor
>
> {code}
> scala> sc.parallelize(Seq(1, 2, 3)).mapPartitions(xs =>
> xs.map(2*).zip(xs)).collect.foreach(println)
> ...
> java.util.NoSuchElementException: next on empty iterator
> {code}
>
>
> Workaround - implement zip with mapping to tuple:
> {code}
> scala> sc.parallelize(Seq(1, 2, 3)).mapPartitions(xs => xs.map(x => (x * 2,
> x))).collect.foreach(println)
> (2,1)
> (4,2)
> (6,3)
> {code}
>
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
