Github user mateiz commented on the pull request:
https://github.com/apache/incubator-spark/pull/468#issuecomment-35179446
Also a second note, related to cleanup -- if you make a single top-level
directory, then cleanup can just be done in the driver program's shutdown hook,
or even better in SparkContext.stop (actually you should do it in both because
the user is not guaranteed to call SparkContext.stop; but if they do call that
you can remove the shutdown hook).
Finally about the lazy init thing, I personally don't like using a lazy
variable if we're confused about the exact semantics and how it interacts with
synchronized. Just manage stuff by hand if it gets confusing, the same way
you'd do it in Java.
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. To do so, please top-post your response.
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.
