Re: [j-devel] Stabilizing for release

[Mark E. <ev...@pa...>]

Erik Huelsmann wrote:
> On Tue, Feb 10, 2009 at 3:28 PM, Ville Voutilainen
> <vil...@gm...> wrote:
>> On Tue, Feb 10, 2009 at 4:07 PM, Erik Huelsmann <eh...@gm...> wrote:
>>> Our progress on compiler-pass1/2 is great, so I wouldn't want to hurt
>>> development speed until we release. These two considerations are
>>> largely conflicting, that's why I want to propose to continue the
>>> destabilizing work, but that this work be performed on a branch
>>> (branches/unstable or something like it).
>> For some reason I'd find it better that we branch the release, and continue
>> working on the svn HEAD as usual. That's what I'm more used to, aka releases
>> are branched, trunk never freezes. We do tag the releases anyway, so we
>> can even branch the release as needed. Maybe do a pre-release tag and if
>> nothing changes for the release, we can just add the release tag to the
>> same version?
> 
> I'm fine with either. From Mark's mail, I understand that branching
> will be the way to go: we expect changes to the lisp based build
> (parity with build.xml) before the final release.
> 

Not all the changes I would propose for 0.13 are in the build system: 
Right now [TRACE is once again broken on trunk][1] that should be fixed 
before release.  And I think there is a rather deep bug somewhere within 
ABCL that breaks SLIME CVS HEAD that I would love to resolve.  But 
abcl-0.12.0 is broken in this way as well (abcl-0.0.11 is the oldest 
version that still works).

The changes to Lisp based build should be completely orthogonal to what 
source level changes are occurring in trunk.

If we branch for release, I would propose we still work on trunk, 
merging the needed changes over to the branch.

[1]:  http://trac.common-lisp.net/armedbear/ticket/47

-- 
"A screaming comes across the sky.  It has happened before, but there is
nothing to compare to it now."