On Fri, 8 Jun 2001 12:00:56 +0200, "Carsten Ziegeler" <[EMAIL PROTECTED]> wrote:
> > Giacomo Pati wrote:
> >
> > There are some drawback to this procedure.
> > 1. Lots of merge conflict will happen because of keyword expansion we use
> > in the source files ($Revision 1.2.1.$ etc.). Using the -kk options
> > on the join command can corrupt binary files if my understanding
> > of that option is correct.
> > 2. The log comments entered on branch commits are not joined into
> > the HEAD
> > branch.
> >
> Yes and because of these two reasons I would suggest that we keep both
> branches in sync which means all commits for the 20 branch should also
> be applied to the 2.1 branch (except the stuff which is really only for
> 2.0). Of course, we should not add all commits of the 2.1 branch to
> the 2.0. So keeping in sync is perhaps not the right formulation.
> I could imagine that it is a very hard work to join the branches after
> some weeks without making mistakes.
I also believe we should be commit both on the 2.0 branch and on the
HEAD(*). Some patches designed for the branch may not apply cleanly to
the HEAD, and in fact may require a slightly different
implementation. It's just a lot easier to apply small patches to the
two branches, as it is done by multiple individuals, than having
somebody come at a later point and doing all the work. The drawback
with this however is that it may increase the time needed to test the
two branches.
(*) Carsten, I assume that the 2.1 branch you mention above is really
the HEAD. There should be no need for a 2.1 branch until we actually
release 2.1.
Regards,
--
Ovidiu Predescu <[EMAIL PROTECTED]>
http://orion.nsr.hp.com/ (inside HP's firewall only)
http://www.geocities.com/SiliconValley/Monitor/7464/ (GNU, Emacs, other stuff)
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
