Subject: Re: the path from nathanw_sa -> newlock2
To: Brett Lymn <firstname.lastname@example.org>
From: Thor Lancelot Simon <email@example.com>
Date: 02/16/2007 07:53:06
On Fri, Feb 16, 2007 at 08:01:59PM +1030, Brett Lymn wrote:
> On Thu, Feb 15, 2007 at 10:56:14AM -0800, Bucky Katz wrote:
> > The whole point of our trying to get the patches into the CVS tree is to
> > reduce the workload of having to continually reapply patches every
> > time we pull a new CVS tree into our work environment. It takes me
> > most of a day, once a month, to merge our work now. The more that goes
> > into the tree we're working on, the less time that merge takes.
> This may be an egg sucking demonstration but you _are_ importing the
> NetBSD sources onto the vendor branch of your repository aren't you?
He's using Perforce, not CVS. I am in a similar situation with FreeBSD
at work: this is basically the only thing Perforce does much worse than
CVS does. It is possible (typically you keep a //vendor in the depot
as an ersatz "vendor branch" and integrate from there to the depot
location where you keep your modified sources) but somehow it always
seems to mean hours with the Perforce merge tool for even trivial
changes, particularly since Perforce seemingly cannot be taught to
ignore any other VCS's keywords.