On Tue, 2008-03-25 at 17:33 +0700, Erik Grinaker wrote:
> On Tue, 2008-03-25 at 15:50 +0530, Bhuvaneswaran Arumugam wrote:
> > On Tue, 2008-03-25 at 15:55 +0700, Erik Grinaker wrote:
> > May I suggest to add "-q" switch for mergeinfo command to address this
> > case? If this switch is passed, we can avoid printing the merged and
> > eligible revisions for a deleted target.
> > If this proposal sounds ok, i'll come up with a patch for this.
> Sounds good to me. I'd prefer this being the default behaviour, but a
> switch is fine.
While I was preparing a patch for this, Kamesh and I found an issue in
current implementation. When a deleted target has unmerged changes in
it, mergeinfo command does not detect it.
For example, consider the attached script. The b1/ is a copy of trunk,
the change r5 in b1/ is not merged to trunk, and b1/ got deleted in r7.
When we run mergeinfo on trunk, it just prints "(source no longer
available in HEAD)" and it does not detect the unmerged change r5 in b1.
Received on 2008-03-25 15:08:33 CET