Front page | perl.perl5.porters |
Postings from May 2012
On Mon, 30 Apr 2012, James E Keenan via RT wrote:
> On Tue Sep 24 07:46:07 2002, doughera wrote:
> > On Tue, 24 Sep 2002, Merijn Broeren wrote:
> >
> > > Quoting Andy Dougherty (doughera@lafayette.edu):
> > > >
> > > > Yes, and it's already done :-). There's an existing $Config{afs} key
> > > > that's either 'true' or 'false' that's designed for such tests.
> > > >
> > > This is the flag to say that AFS is present on systems where this
> > > Perl is likely to run.
> >
> > Ah, that's a tricky distinction, and one which Perl's Configure system is
> > not well-set-up to handle (and which I didn't pick up at all). I mean the
> > distiction between the "system where perl is built" and "the system where
> > perl will run". A very common place that distinction would be relevant is
> > when perl is supplied by a vendor (e.g. Sun) that is compiled with the
> > vendor's compiler, but is likely to be installed on systems that only have
> > gcc. But that's a separate Configure "Todo" item.
> >
> > > Since I supplied the original incomplete patch, I shall have a look for
> > > all these kind of tests in the current configure and install scripts and
> > > then provide a next patch. Which needs to be carefully checked :-)
> >
> > Thanks!
> >
> > Andy Dougherty doughera@lafayette.edu
> >
>
> Do we know if we still have these configuration issues on AFS?
Yes, we do. Configure still uses a plain "afs" in spots where it should
probably use $afsroot. It's probably not a high priority problem,
however. I'd mark it as "stalled".
--
Andy Dougherty doughera@lafayette.edu
Thread Previous
|
Thread Next