[...]
> > Hmmm, this would imply that hg clone clones unresolved (or) files. I
> > don't think this is the case (or am I wrong?). I would think that
> > children will have to re-parent to master workspace and re-pull from
> > master workspace. But I haven't tried that.
>
> If the clone doesn't clone the unresolved merges, that's fine. The
> solution is (as you suggest) for the child to pull that file from the
> same two sources (reference copies of the project gate and the master
> gate) and replicate the same conflict.
>
> As long as the merge results and proper metadata markings ("these
> files were merged at this point") make it into the merge workspace, I
> don't think it matters too much how hard it is to accomplish.
>
> It'd be nice to see the clone just "do the right thing," but I can see
> how that might be hard.
>
> (At least in teamware, that doesn't involve actual 'reparenting', as
> you can specify to pull from _any_ workspace.)
Hmm, looking at 'hg help pull', Mercurial is supposed to do that also.
One last thing which came to my mind. If you do several merges at "merge
workspace", you will want to do final putback to "master workspace".
For this might be nice to collapse all the submerges into one changeset.
hg diff -r BASE -r tip > merge.diff
hg strip -r BASE+1
gpathch < merge.diff
hg ci ...
hg push ...
--
Vlad
pgpadAWwz5Ik1.pgp
Description: PGP signature
_______________________________________________ tools-discuss mailing list tools-discuss@opensolaris.org
