Re: coywolf's gnumach lxr

[Marcus Brinkmann]

At Thu, 23 Dec 2004 12:17:43 +0800,
Coywolf Qi Hunt <address@hidden> wrote:
> On Wed, 22 Dec 2004 15:50:56 +0100, Marcus Brinkmann
> <address@hidden> wrote:
> > At Tue, 21 Dec 2004 23:15:23 +0800,
> > Coywolf Qi Hunt <address@hidden> wrote:
> > >
> > > hello,
> > >
> > > I've just made this gnumach lxr.  I was looking for hurd lxr. And the
> > > people in #hurd encouraged me to set up one myself. So, this is it.
> > > http://sosdg.org/~coywolf/lxr-gnumach/source/ Should it be added to
> > > the offical hurd page?
> > 
> > How is this different from
> > 
> > http://www.htu.tugraz.at/~past/hurd/global/
> > 
> > which already exists for a long time now?  Can you point out
> > differences (in particular functional ones)?
> 
> I knew gnu global and that link. But i am familiar with lxr. If something
> goes wrong, i can fix the lxr perl script.

That's fine, but what are the functional differences?  We can hardly
advertise it on the web page as the "cross reference based on the
cross reference generator which coywolf is familiar with".  Ie, if we
offer two links to two different cross references, we should provide
the visitor of the web page with enough information to make a sensible
decision which one to pick.

> > If you are committed to keep the site up at that URL for a long term,
> 
> Yes, I will. And I have Brian Bruns' full support. He is one of the
> sosdg admins.

Ok, great.

> But maybe in future i'll have to change the url 
> http://sosdg.org/~coywolf/lxr-gnumach/source/
> to 
> http://sosdg.org/~coywolf/lxr-foo/source/
> or something similar, or just setup another link for hurd source.
> 
> Since hurd isn't released, i don't set up hurd source yet.

Actually, the Hurd is released (0.2 was released 1997), so if that's
the only thing holding you up, you should add it, and mig, too.

It would also make much sense to add glibc, as a bunch of the Hurd's
functionality is implemented there, and it cross references heavily
with the Hurd (although not always in machine-parseable ways I am
afraid).

Thanks,
Marcus