[Catalyst] [Announce] Catalyst-Runtime 5.8000_04 shipped to CPAN

Guillermo Roditi groditi at gmail.com
Tue Dec 9 12:38:51 GMT 2008


On Sun, Dec 7, 2008 at 1:47 PM, Tomas Doran <bobtfish at bobtfish.net> wrote:

>
> I'm not sure if the correct fix for the lies in Catalyst, or MX::E::CAF,
> but:
>
> http://dev.catalystframework.org/svnweb/Catalyst/revision?rev=3D8781
>
> is the minimal test case I've been able to produce for the breakage of the
> Catalyst::Plugin::Authentication's tests and which people still using the
> back-compat auth plugins in their apps would be seeing (if any of them we=
re
> trying it and shouting).
>

I think it's the fact that the mk_accessor call is in BEGIN {} because the
adopt happens at compile time, but the role application within it happens at
runtime.

I am giving this a wontfix until I can figure out if it is even fixable. I'm
not giving up on you, though. Why is it necessary to have the accessors
created at BEGIN time?

I'll experiment with having the role consumption happen in a BEGIN block,
but I don't know what the deal will be with it. I'll keep you all posted.

-- =

Guillermo Roditi (groditi)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.scsys.co.uk/pipermail/catalyst/attachments/20081209/f9406=
784/attachment.htm


More information about the Catalyst mailing list