cloud-fan commented on a change in pull request #28983:
URL: https://github.com/apache/spark/pull/28983#discussion_r449465235
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/objects/objects.scala
##########
@@ -679,7 +679,10 @@ object MapObjects {
elementNullable: Boolean = true,
customCollectionCls: Option[Class[_]] = None): MapObjects = {
val loopVar = LambdaVariable("MapObject", elementType, elementNullable)
- MapObjects(loopVar, function(loopVar), inputData, customCollectionCls)
+ // println(s"\n\n $function, $inputData, $elementType, $elementNullable,
$customCollectionCls \n")
+ val fOfLV = if (function == null) loopVar else function(loopVar)
Review comment:
Yea this is the core issue: we don't expect to serialize and send
`UnresolvedMapObjects` to executors. Sometimes users may carry it if the RDD
closure is not carefully written. We allow this mistake but we can't allow if
users do want to process the `UnresolvedMapObjects` at executor side.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
