Re: make web fails

[Erik Sandberg]

On 6/2/06, Han-Wen Nienhuys <address@hidden> wrote:
> Erik Sandberg schreef:
> > It seems to work equally well as CVS HEAD, so it doesn't break anything (it's
> > just that I hack code all the time, and I don't like to have too big or many
> > development forks active at the same time). The code doesn't have gs problem,
> > it's likely my gs installation that has problems.
> >
> > FWIW, my previous patch for CreateContext has grown over time, and now
> > contains implementations of RemoveContext, ChangeContext and Prepare events
> > as well, and my parser patch has also grown considerably. I think it would be
> > much easier for me to develop if I could commit _some_ stuff before I get GUB
> > running (e.g., it would be good if I could commit fixes for some of the
> > issues you brought up below).
>
> I understand that keeping several branches in the air is a pain,
> especially because CVS makes it so awkward.  However, it's something
> that should be solved on your side of the fence.  I don't mind working
> together on a technical solution that makes it easier for you to have
> more branches, but keeping HEAD in working order is more important, and
> part of that is to make sure that we don't introduce new problems, and
> fix old problems before adding new stuff.

I have tried to get gub working today, and there are problems.
Furthermore, I won't have a good Internet connection until wednesday,
so I suspect it will take some time before gub works.

Can I commit bugfixes for those old problems by sending patches to you
meanwhile, or do you prefer to wait?

(current gub problem: because of the 64-bit problem, guile required
two makes to build, so gub-builder now barfs and wants me to pass
--force-package, but I can't find where to add that flag)

BTW, will gub need to download much more from internet after make
bootstrap succeeds?

Erik