On 9 Dec 2008, at 12:46, Guillermo Roditi wrote:
On Sun, Dec 7, 2008 at 4:38 PM, Tomas Doran <[EMAIL PROTECTED]> wrote:Please find attached a simple test case for the behavior needed by Catalyst::Plugin::Cache::Curried (and anything else which says __PACKAGE__->mk_accessors(qw/ meta /)) - found by looking at MojoMojo's current test failures.Yeah dude. wontfix. You don't get to take over Moose's "meta" method. Immutable replaces the accessor when inlining the meta method. You can argue with Moose over who has the right to that particular method name, but that's out of the scope of my module.
No it doesn't (immutable)?I agree that all 'real' Moose classes shouldn't be breaking the meta method like this, but the fact is that there *IS* real code out there which is going to be broken by this, and it's something which we can _avoid_ breaking without too much effort..
Sure, we should be warning people, but breaking their applications when we don't _need_ to just seems silly.
Attached nasty hackery which makes my initial test pass, proving that it is possible..
Cheers t0m
meta-method.diff2
Description: Binary data
_______________________________________________ List: [email protected] Listinfo: http://lists.scsys.co.uk/cgi-bin/mailman/listinfo/catalyst Searchable archive: http://www.mail-archive.com/[email protected]/ Dev site: http://dev.catalyst.perl.org/
