On Tue, Jan 27, 2015 at 12:06 PM, Belcourt, Kenneth <kbelco@sandia.gov> wrote:

On Jan 21, 2015, at 2:42 PM, Belcourt, K. Noel <kbelco@sandia.gov> wrote:

> On Jan 21, 2015, at 2:30 PM, Rene Rivera <grafikrobot@gmail.com> wrote:
>
>> FYI.. I've started restructuring the regression repo to better know what the various tools are and the subparts thereof. And to try and work towards faster, cleaner, and easier to maintain code. And part of that restructure is done enough for report generation use. So..
>>
>> On Fri, Jan 16, 2015 at 11:00 AM, Rene Rivera <grafikrobot@gmail.com> wrote:
>> I'm going to remove the outdated regression testing code in the boostorg/boost super-repo the weekend of January 24th. Everyone MUST stop using the old scripts immediately as they will disappear in approximately 7 days.
>
> While I fully appreciate and support your desire to do this, it appears there’s something wrong with the new version, at least our uses of it.  I’ve tried several things and our testers do not seem to run correctly under the new run.py, no idea why as yet.  Could I ask that you please not remove the old stuff until we get this issue resolved?

Alright, I’ve found the source of our problem with the new run.py script.  For whatever reason, PJL was being built only once, rather than once per tester.  I guess this is because I used to run testers with a completely clean test directory so PJL was built each time a tester ran.  With the new run.py script, it was finding a previously built PJL but was unable to locate the dependent shared libraries due to an overwritten LD_LIB path.

So Rene, I think you can safely remove the outdated regression testing code, thanks for giving me a chance to track this down.

Thank you for tracking it down quickly! I've scheduled the deletion for this weekend.


--
-- Rene Rivera
-- Grafik - Don't Assume Anything
-- Robot Dreams - http://robot-dreams.net
-- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail