On Thu, Feb 14, 2013 at 11:34 AM, Bill Moseley <mose...@hank.org> wrote:
> > > On Thu, Feb 14, 2013 at 11:13 AM, Jesse Luehrs <d...@tozt.net> wrote: > >> > Baseclass->meta->remove_method( 'BUILD' ); >> >> No, that's almost certainly a bad idea. I think the best option in this >> case is to take this question to #catalyst or #dbix-class, and see if >> they can come up with a better solution(: >> > > How bad? :) Like bad break or bad ugly? I think the only other solution > is to build my own version of Catalyst::Model::DBIC::Schema. > Sorry to poke on this, but in a meeting yesterday afternoon discussing solutions other developers asked this same question. Mostly wondering if it can potentially break Moose or it was simply that brute-force removal of a parent's method is simply a bad idea in general -- especially one that the parent's author might have assumed would always run. It would be great if I could give them a more "official" answer. In the end we replaced Catalyst::Model::DBIC::Schema with our own modified version. But, that's about as brute-force replacement of the parent's BUILD method as you can get. :) But, there could be cases where one would not want to do something so global, yet need to override or somehow modify the BUILD method as one might normally do for other methods via a subclass. Thanks, -- Bill Moseley mose...@hank.org
