What are the project deps that really block or hold TomEE back at the
moment?
On 08/04/2014 08:28, Romain Manni-Bucau wrote:
little ps: it is not enough ATM for openjpa to use asm5 shade to work
on java8, more work is needed. I'll try to have a look but it would
surely need to rewrite the enhancing/proxying with asm.
Romain Manni-Bucau
Twitter: @rmannibucau
Blog: http://rmannibucau.wordpress.com/
LinkedIn: http://fr.linkedin.com/in/rmannibucau
Github: https://github.com/rmannibucau
2014-04-07 22:57 GMT+02:00 Romain Manni-Bucau <rmannibu...@gmail.com>:
well something which is doable today (actually already "demo" it on
current version) is batchee which is now released and usable.
Trying so summarize:
1) jbatch -> just to import
2) websocket -> out of the box
3) CDI -> OWB to hack before integrating
4) BVal -> model to build in tomee, cdi integration to rework a bit
(needs to exclude bval scanning) but work startable
6) jaxws/jaxrs -> cxf almost done but not yet and api not yet the
right one last time i looked so blocking
7) openjpa and amq -> not yet started
Romain Manni-Bucau
Twitter: @rmannibucau
Blog: http://rmannibucau.wordpress.com/
LinkedIn: http://fr.linkedin.com/in/rmannibucau
Github: https://github.com/rmannibucau
2014-04-07 22:54 GMT+02:00 Jean-Louis MONTEIRO <jeano...@gmail.com>:
Oh yes, definitely agree on side effects.
In the roadmap, i did not mentioned precisely the numbers.
I'm pretty sure for the 1.6.0.1 and the 1.7.0 because it's done.
Then, it's not a short term. But I like to start before June. Of course and
as you mentioned it depends how far and how much we can work on other
projects. Let's start small with BVal.
Then, OpenWebBeans should be ok and we can finish with OpenJPA.
JLouis
2014-04-07 22:50 GMT+02:00 Romain Manni-Bucau <rmannibu...@gmail.com>:
why I'm not so hppy to get 2.0 now is we'll surely fork at a point
with 1.7.x branch. Can be a lot short term since other projects are
really not ready to be integrated yet. I expect few months before
being able to do it constructively. Can be done in parallel surely but
not breaking tomee is an important challenge. We need to avoid what we
did for 1.0.0 where trunk was broken for a while and in between
releases/snapshoting was really hard...can also mean we want to use a
real git repo now to be able to switch faster between
versions/branches (today switching between versions/branches is really
slow if you don't use vim).
Romain Manni-Bucau
Twitter: @rmannibucau
Blog: http://rmannibucau.wordpress.com/
LinkedIn: http://fr.linkedin.com/in/rmannibucau
Github: https://github.com/rmannibucau
2014-04-07 22:42 GMT+02:00 Jean-Louis MONTEIRO <jeano...@gmail.com>:
1.6.1: totally opened question, much more for users in order to have a
better compatibility insurance ;-) Not sure going to 1.7.0 in that scope
is
more risky, but the question makes sense at least, whatever the result
is.
2.0: right, having a branch would help us investigate and try to create
patch for other projects, like you mentioned. Of course to integrate all
Java EE 7 compatible projects we need at least to have something. But a
branch allow us, like in the past to fork locally, patch to reverse back
to
the community. It's more for OpenJPA, BVal and MyFaces we don't have that
much people involved.
The in between step is ok for me, if that makes sense.
Thanks for the accurate feedback
JLouis
2014-04-07 22:29 GMT+02:00 Romain Manni-Bucau <rmannibu...@gmail.com>:
+1 for 1.6.0.1
+1 for 1.7.0 with java 8 support (side not is all patches are applied
on respectives trunk since today so we can "just" fork but dont think
we need) when possible (openjpa is on the red path so we should just
ping them, otherwise releases are here or can be here in less than a
week)
1.6.1 is useless IMO
2.0 when 1.7 is released is possible but the main point about it is
IMO we should make other project released first (bval is ready, owb
needs some work) otherwise we'll don't get much benefit to use their
snapshots but a lot of drawbacks (+ bval snapshot is not really here
actually). So I'd start JavaEE 7 by other project and would hack it up
in tomee when ready to integrate sthg (bval seems the first real
candidate). This would allow a last fixes release before JavaEE 7 is
needed or just wouldn't have had any drawback if not. A in between
step would be to pass trunk on java 7 and drop java 6 support without
adding EE 7 feature (a mandatory first step btw). wdyt?
Romain Manni-Bucau
Twitter: @rmannibucau
Blog: http://rmannibucau.wordpress.com/
LinkedIn: http://fr.linkedin.com/in/rmannibucau
Github: https://github.com/rmannibucau
2014-04-07 21:59 GMT+02:00 Jean-Louis MONTEIRO <jeano...@gmail.com>:
Hi devs,
now that the security maintenance release 1.6.0.1 is ongoing, time to
look
in the upcoming months. Here are some thoughts on a possible roadmap.
- Apache TomEE 1.6.0.1 (now)
Security maintenance release 1.6.0.1 --> Cf. previous mails on the
dev@list.
- Apache TomEE 1.7.0 (next week, right after the 1.6.0.1)
This is a feature full release with some major parts: Java 8 support
(only
2 weeks after the GA, Apache TomEE plume (Apache TomEE Plus + Mojora +
EclipseLink) to name a few.
We are almost ready to release, but some dependencies need to be
upgraded
(CXF, XBean, OpenJPA, OpenWebBeans).
We can do that by applying the patch internally on our trunk and
deliver
our own binaries.
The good point of such an approach is that we can gather as many
feedback
as possible quickly and then we have room to spread the word over
other
communities.
- Apache TomEE 1.6.1 (??)
That releas could be basically the current trunk, without java 8
support
and without the plume distro.
WDYT?
Any interest on the community side?
If yes, we could do this one right after the 1.7.0.
- Apache TomEE 2.0 branches
One month after the 1.7.0 we could start the 2.0 branch to target Java
EE 7.
That branch gonna be the new trunk and we'll keep active branches for
1.6
and 1.7.
That means that all fixes mergeable to those branches must be done on
the
flow.
WDYT?
Any feedback welcome.
--
Jean-Louis
--
Jean-Louis
--
Jean-Louis
--
Andy Gumbrecht
http://www.tomitribe.com
agumbre...@tomitribe.com
https://twitter.com/AndyGeeDe
TomEE treibt Tomitribe! | http://tomee.apache.org
