DBIx::Class and minimum perl requirements (especially EPEL)

Nigel Metheringham nigel.metheringham at dev.intechnology.co.uk
Tue Oct 28 17:24:16 UTC 2008


On 28 Oct 2008, at 17:04, Chris Weyl wrote:

> On a F-8 system with the latest perl (perl-5.8.8-41fc8), the first
> test failed, but the second test passes.  This makes me worry, as
> we're failing a test here on an F-8 system that doesn't actually have
> this problem.

Its changed since then :-)

Theres been quite a lot of to and fro discussion in the last 24 hours  
on the best way of doing things, and its changed from my original  
implementation (a few times actually).

I haven't been testing on F8, but have on Centos 5.2 systems with both  
the original (buggy) perl and the September update (not buggy).  I  
would expect them to behave just like F8.

A fixed perl now gives no errors, the C5.2 original perl fails.

> Given that these tests look for something that is specifically a
> RH/Fedora issue, would it be unreasonable to look for specific
> Fedora/RH information in determining the first test pass/fail?

Maybe - except I'd like to mop up any downstream distributions too, so  
not *too* RH/Fedora specific :-)

Would you care to try the current trunk version (its merged now so the  
branch is obsolete):-
   http://dev.catalyst.perl.org/repos/bast/DBIx-Class/0.08/trunk

or just grab the file itself (it only needs perl core modules to run):-
   http://dev.catalyst.perl.org/svnweb/bast/view/DBIx-Class/0.08/trunk/t/99rh_perl_perf_bug.t

[And, hey, there have been no changes to this for 4 hours now - thats  
just about counting as stable].  More seriously we think that this is  
usable now, although when the dev snapshot hits CPAN I won't be  
suprised to find people with different opinions.

We now also work with 5.8.9 dev versions.  The main thing I'd change  
given copious spare time is removing the RH labelling from the test  
code as its really testing for a 5.8.9 behavioural change that bless  
affects references rather than the thing referred to.

	Nigel.
--
[ Nigel Metheringham             Nigel.Metheringham at InTechnology.com ]
[ - Comments in this message are my own and not ITO opinion/policy - ]




More information about the Fedora-perl-devel-list mailing list