also sprach Manoj Srivastava <sriva...@acm.org> [2009.05.10.0028 +0200]:
> > Keep in mind that the branches accumulate remotely, so you'd have to
> > pass -r or -a to git-branch to be exposed to them.
> 
>         Accumulate?
[...]
> > Anyway, unfortunately, published branches cannot be renamed with
> > Git, or retired, or anything of that sort, so the concern is valid:
> 
>         Why not?

Because they might exist in downstream repos and could be pushed
back at any time. However, contrary to my initial believe, this
requires explicit pushing with standard Git. TopGit currently,
however, overrides the default push refspec and effectively defaults
git-push to --all.

>  git checkout     topic--obsolete-later
>  git checkout -b  old-obsolete-now
>  git push --all
> 
>  git branch -d topic--obsolete-later
>  git push origin :topic--obsolete-later
>  git remote prune origin

Right, until I push back topic--obsolete-later, which I had
previously checked out.

> > you'd have to start a new feature branch as obsolete/new-feature
> > to be able to declare it obsolete later on, at which point we
> > might just as well not use the namespace. :)
> 
>         This sounds wrong to me. What am I missing?

Mainly my list mail following this one. :)

-- 
 .''`.   martin f. krafft <madd...@d.o>      Related projects:
: :'  :  proud Debian developer               http://debiansystem.info
`. `'`   http://people.debian.org/~madduck    http://vcs-pkg.org
  `-  Debian - when you have better things to do than fixing systems
 
wind catches lily,
scattering petals to the ground.
segmentation fault.

Attachment: digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/)

_______________________________________________
vcs-pkg-discuss mailing list
vcs-pkg-discuss@lists.alioth.debian.org
http://lists.alioth.debian.org/mailman/listinfo/vcs-pkg-discuss

Reply via email to