Github user adamlamar commented on the issue:
https://github.com/apache/nifi/pull/2361
> Do we risk making duplication by updating currentKeys in the middle of
the loop?
Yes, I think we do! I identified a similar (possibly the same) bug, and I
agree with all of your suggestions. The question in my mind is whether we
should fix all of these issues in this JIRA or defer to another. As far as the
original JIRA goes, I believe the current commit will address the issue. I also
did a fair bit of manual testing so I would be comfortable moving forward with
this change as-is.
Before refactoring, I'd like to put some additional unit tests in place for
safety. Its clear from the discussion that there is some meat here and I'd
really like to enumerate a few cases I've seen while testing in unit tests.
So its up to you - would you prefer that I start the unit tests and address
(potentially) multiple bugs in this PR, or should we merge this and create
another JIRA?
---
