[cle-release-team] JUnit upgrade to at least 4.11?

[Matthew Buckett]

Assuming everything passes it seems a no brainer.

One issue I've seen on local projects is:

https://jira.springsource.org/browse/SPR-5145

which has meant locally we couldn't move past jUnit 4.4 until we
upgrade to spring 3. But it may wel be that Sakai doesn't have this
issue.

On Mon, Apr 15, 2013 at 11:45 PM, Matthew Jones <matthew at longsight.com> wrote:
> JUnit 4.11 is the latest version released.
>
> It brings a pretty big improvement with JVM 7; more reliable tests and a
> predictable method orders. There are 3 options
> DEFAULT -  Deterministic order based on an internal comparator but not
> predictable. (At least this means if it passes once, future failures aren't
> because of the order, unless some new test is added?)
> JVM - Order may vary from run to run
> NAME_ASCENDING - Based on method alphabetic names
>
> "Of course, well-written test code would not assume any order, but some
> does, and a predictable failure is better than a random failure on certain
> platforms."
>
> https://github.com/junit-team/junit/blob/master/doc/ReleaseNotes4.11.md"
>
> In kernel, junit is currently 4.8.2, in master junit is 3.8.1. Seems like
> these could both be upgraded. It fixed a "seemingly random" test problem I
> was seeing recently. I did this locally and they built without problems
> (though that was expected) and the tool (dashboard) I was having random
> failures with has been building reliably with this version upgraded locally,
> was about 50% failure before.
>
> _______________________________________________
> cle-release-team mailing list
> cle-release-team at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/cle-release-team
>



-- 
  Matthew Buckett, VLE Developer, IT Services, University of Oxford