Github user mateiz commented on the pull request:
https://github.com/apache/spark/pull/136#issuecomment-37879303
BTW the way I would write this is that each partition's compute() should
return all the windows that start with elements in that partition. To do this
it may have to read ahead to compute more partitions of the RDD (usually at
least one, and sometimes even more) to close off the windows for its last few
elements. If you're not careful though this can be very expensive, for a sparse
RDD and a large window size, so I'm still not 100% sure we want to do it.
---
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.
---
