rednaxelafx commented on a change in pull request #28463:
URL: https://github.com/apache/spark/pull/28463#discussion_r422033702
##########
File path: core/src/main/scala/org/apache/spark/util/ClosureCleaner.scala
##########
@@ -256,7 +226,7 @@ private[spark] object ClosureCleaner extends Logging {
return
}
- if (lambdaFunc.isEmpty) {
+ if (maybeIndylambdaProxy.isEmpty) {
logDebug(s"+++ Cleaning closure $func (${func.getClass.getName}) +++")
Review comment:
I left this part untouched in this PR intentionally, for the reason
mentioned in a previous reply.
It's highly unlikely to actually see this log get triggered in Spark 3.0 in
the real world, unless someone decided to go with `-Ydelambdafy:inline`
explicitly w/ Scala 2.12+. So keeping it the way it is for Spark 3.0 wouldn't
hurt much.
I do plan to send out a few cleanup/refactoring PRs to get the old code into
a better shape, and maybe enhance the ClosureCleaner's capabilities, e.g.
support cleaning "sibling" closures.
----------------------------------------------------------------
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]
