Re: Race condition merging branches

[Carel Fellinger]

On Mon, Aug 27, 2001 at 01:45:03PM -0400, Eric Siegerman wrote:
> On Mon, Aug 27, 2001 at 03:15:10PM +0200, Carel Fellinger wrote:
> > I naively thought that the
> > follwing would do:
> > 
> >   cvs rtag -r branch branch_n
> >   cvs update -j branch_n-1 -j branch_n
> > 
> > but this obviously leaves out the stuff in branch_n [...]
> 
> This is wrong.  The stuff in the revisions labelled with the
> "branch_n" tag is included in the merge.
> 
> What's excluded is the changes in the "branch_n-1"-labelled
> revisions.  But that's what one wants; they were merged last
> time, having been included in the:
>       cvs update -j branch_n-2 -j branch_n-1
> merge.

Thanks for clearing this up.  The other day it seemed not to work like
this over here. But today I can't get it to malfunction again, so I
must have misinterpreted what I saw.  From trying now I think I got
alarmed by some .#something files differing in the two sandboxes I
used to test the two path to merging the branch, not noting that those
files had little relevance to the end result.

-- 
groetjes, carel