On Wed, 2008-03-12 at 15:06 +0000, John A Meinel wrote:
> Something does seem a little fishy here.
>
> It seems like maybe doing a "bzr record" first would make sure that the
> loom state is set on disk and then the pull wouldn't overwrite it.
>
> Regardless, it would seem that "bzr pull" shouldn't smash the target
> loom record. Just like we do now with file contents, 'pull' preserves
> local modifications.
Definitely a bug; it should set the basis loom to the pulled one, and
the current loom be 3-way merged with the changes from basis to new
basis. Basically the same as for working trees, and I would expect this
to come when 'merge' really merges the looms.
On Wed, 2008-03-12 at 15:06 +0000, John A Meinel wrote:
> Something does seem a little fishy here.
>
> It seems like maybe doing a "bzr record" first would make sure that the
> loom state is set on disk and then the pull wouldn't overwrite it.
>
> Regardless, it would seem that "bzr pull" shouldn't smash the target
> loom record. Just like we do now with file contents, 'pull' preserves
> local modifications.
Definitely a bug; it should set the basis loom to the pulled one, and
the current loom be 3-way merged with the changes from basis to new
basis. Basically the same as for working trees, and I would expect this
to come when 'merge' really merges the looms.
-Rob www.robertcolli ns.net/ keys.txt>.
--
GPG key available at: <http://