Hi Pierre,
This is not orthodox, but after being surprised by the idea of not releasing
R14, I think it make as much sense as not releasing R13.
We could then focus entirely on R15 and I'd even ask if we could not backport
the files moves, now only in trunk, to R15.
So we could then easily backport the changes in those files. It seems those moves went well, which is actually not surprising since apart typos we
could not expect much issues.
Another possibility would be to soon freeze a R16 branch, once the Gradle use
will have been validated for instance.
Opinions?
Jacques
Le 28/06/2016 à 20:09, Pierre Smits a écrit :
+1 on putting the r13.07 branch out to the pasture, and bring the focus
towards r15 branch to cut a release soon.
+1 on putting the r14 branch out to the pasture, and bring the focus
towards r15 branch to cut a release soon.
+1 on focusing on getting the release from r15 out ASAP
Best regards,
Pierre Smits
ORRTIZ.COM <http://www.orrtiz.com>
OFBiz based solutions & services
OFBiz Extensions Marketplace
http://oem.ofbizci.net/oci-2/
On Tue, Jun 28, 2016 at 7:32 PM, Todd Thorner <[email protected]>
wrote:
Thanks for this, Mr. Foga. The Downloads page of the ofbiz.apache.org
site mentions 14.x and 15.x only as part of the project's tentative release
schedule. Would end user newcomers be wise to wait out the transition?
On 16-06-28 09:55 AM, Sharan Foga wrote:
Thanks Jacopo for the details and summary.
I know that some people might think it a bit strange that this discussion
is happening on the user list rather than the dev list, but I think these
topics are something that our users may have an opinion on and want to
comment.
+1 for suggestion #1
13.07 has been around a while so I think that ending it now rather than
next year will help us to focus on other work (e.g. refactoring and clean
up)
+1 for suggestion #2
I think the work we are currently doing to improve the trunk build system
isn't really that visible to our users, but it will make a big difference
to developers and people who administer the system. Pushing those technical
benefits back to 14.12 and 15.12 (if we still want to call them that!) will
help.
Also OFBiz has changed a lot since the 14.12 and even 15.12 branches were
created. We've incorporated so much new functionality that it would be a
good thing to try and get some of this into our next release. Essentially
it would mean more functionality for our users.
Thanks
Sharan
On 2016-06-28 12:26 (+0200), Jacopo Cappellato <
[email protected]> wrote:
Hi all,
as you may know we are working at migrating the build scripts of the
OFBiz
trunk from Ant to Gradle.
Together with this important change we are also modifying, for policy
reasons, the way we distribute the external dependencies (i.e., the jar
files needed by OFBiz): the required jars will be downloaded at build
time.
Since these changes are not bug fixes, the current plan is to do these
changes only in the trunk and do not backport them to the active
branches,
that are currently:
* 13.07
* 14.12
* 15.12
However, we will still have to modify these branches by removing the
external jar files and download them using Ivy.
The first concern is that we will have to work on and stabilize two
fronts:
Ivy for the 3 current release branches and Gradle for the trunk and the
future branches.
The second concern is that, as a consequence, we will have, for several
years, significant differences in the setup/build steps between the old
releases and the new ones that could cause confusion and regressions when
bugs are backported.
We have already issued 3 releases from the 13.07 branch and we have a
tentative plan to issue one more release around February 2017 that would
be
the last release of this series.
As regards 14.12 and 15.12 branches, no releases have been issued yet.
Based on these details I would like you to consider the following
decisions:
1) anticipate the end of life of the release branch 13.07 at now; we
would
not issue the fourth release as initially planned
2) once stabilized, backport to 14.12 and 15.12 all the changes required
to
build the system and download its dependencies with Gradle
Please express your opinion on each of them separately, since they are
independent (i.e., you could agree/disagree on the first/second/both).
Thanks,
Jacopo
