Never mind. Thanks to Brian
de Alwis the problem has been
determined to be in our Unit tests (and deep, private code that is no longer
I will be working on promoting our final
official builds over the next couple of hours.
David M Williams/Raleigh/IBM@IBMUS To:
01/29/2016 01:40 PM Subject:
Eclipse Platform's final, official contribution to Neon M5 and Mars.2 RC2
will be a day late Sent by:
The Platform's final, official contribution to the Sim. Release will be
a day late.
Our current candidates are M20160128-1800 (for Mars) and I20160128-2000
(for Neon). Those doing their own builds for these check points should
be able to safely use those while doing your own "compiles".
Our issue is more related to which EMF version to use and package with
The issue is described in bug 486804 . The issue is still be investigated,
but as of now, we do not know if it is a regression in EMF, or a "test-only"
problem in our Unit tests. And, in the worst case, even if our tests were
making some incorrect assumption related to 'order' or iteraters, we might
be making the same incorrect assumption in our runtime code? While it continues
to be investigated, we are going to do a "back up" build that
has the latest EMF version reverted, just in case that turns out to be
the better version to use at this point in time. The goal is that by the
time that build is ready (or before :) we will understand the root problem
completely. But in any case, we do not expect the issue to be resolved until very late
today or tomorrow, on Saturday.