Sorry for the top post; I'm stuck on Blackberry for the moment.
What I find annoying is when a module correctly specifies its dependencies, but
one of its dependencies fails to specify them correctly. I think in such a case
I would want a report so that I can pursue my dependency's author, or add the
missing dependency to my own PMPREQ until the dependency's author is able to
become responsive.
------Original Message------
From: John M. Gamble
To: Shmuel Fomberg
Cc: module-authors
Subject: Re: CPAN testers: failure installing dependencies
Sent: Nov 5, 2012 9:04 AM
On Mon, November 5, 2012 1:54 am, Shmuel Fomberg wrote:
> Hi All.
>
> Can anyone tell me, when a cpan smoker / tester is trying to test a module
> but fails to install a dependency, what happens?
>
> Who should I convince to make such a failure similar to test fail?
> Or maybe mark it as failure to all the modules that depend on the failed
> module?
>
> I seek to do it because lately modules feel free to have deep
> dependencies,
> a lot of modules have some small test failure percentage, and these add up
> rather quickly.
>
> It is annoying to try and installed some new-shiny module, just to see
> some
> dep^3 fails testing, and then to start investigating.
>
> Thanks,
> Shmuel.
>
Sorry for the duplicate message Shmuel, I forgot to hit "Reply All".
I'd count it as a bug, and file a report that way.
For example, if their module depended on Math::Complex, and they failed to
set up the dependency in (say) their Build.PL, I'd file a bug report
saying that they needed to add the line
'Math::Complex'=> '1.36', # Bugfix of Im().
to their "requires" hash.
-john
-- Dave Oswald
