Re: [Gnu-arch-users] Re: --skip-present implemented

[Miles Bader]

On Tue, Sep 09, 2003 at 05:33:56AM -0700, Tom Lord wrote:
> That's the only caveat.   You _say_ you do pure merges -- and you
> probably do -- but you'll have to maintain a discipline to be sure
> that that's _really_ what you do.

Yeah, I'm aware of the issue, and have certainly had various conflicts result
in not-quite-pure-merges, but have had no real problems so far.

In general it's my experience that since most conflicts that require sullying
a `pure merge' come from differences between the trees, and if I do a back
merge that skips the aforementioned sullied-pure-merge, and the difference is
propagated into the other tree in another changeset, it results in the same
conflict `in reverse.'  So things sort of even out -- the conflict resolution
is in different changesets in the two branches, but that makes sense given
that some changesets occur in different orders.

> And if you toss in some appropriate sync-trees, there, or if the
> merges thus performed happens to give you the latest patch-logs in
> main and in the branches, then star-merge will be happy again.

I kind of wondered if it would good/bad/optional for replay --skip-present
to actually add the patch-logs from skipped revisions (but regardless of
desirability, it would probably be a slight pain to implement).

-Miles
-- 
Suburbia: where they tear out the trees and then name streets after them.