Front page | perl.perl5.porters |
Postings from January 2012
On Fri, 27 Jan 2012, David Golden wrote:
> On Fri, Jan 27, 2012 at 2:04 PM, Andy Dougherty <doughera@lafayette.edu> wrote:
> >
> >> (b), I don't think we have to automate a solution, only tell the user
> >> that they're doing it wrong.
> >
> > Actually, no. I keep the perl sources on an NFS-mounted directory. I use
> > one system (that has git) to update the sources, but I'm often building,
> > using the same directory, on a different system that doesn't have git.
>
> I would define that as one variation of "doing it wrong" and I don't
> think it's the responsibility of the core to save you from yourself,
> since hybrid git/NFS-mounted multi-system build directories aren't
> exactly the norm. (Particularly since you're well qualified to know
> what risks you're taking with a setup like that.)
I fear I didn't express myself clearly. I meant "no" in response to your
statement
> Per your identification of the problem, we can assume (a).
(i.e. git is present).
I completely agree with (b), and saw Nicholas's approach as a good way to
"tell the user that they're doing it wrong."
> That said, I think Nicholas made a good point for why the extra
> (non-git-based) check is useful.
Yes, I think we're all on the same page here.
--
Andy Dougherty doughera@lafayette.edu
Thread Previous
|
Thread Next