On Tue, Sep 30, 2008 at 3:12 AM, Jeff Hodges <[EMAIL PROTECTED]> wrote: > Seth Thomas Rasmussen wrote: >> >> Back on point, I dunno if throwing a patch like this in right away >> would be best if some other namespace related changes weren't made. I >> kind of want _why to weigh in on my question from the other thread >> about how necessary and flexible the auto-namespacing behavior we have >> now is. >> >> I think Bluebie mentioned how the current behavior is similar to how >> Camping does things. I remember bumping into that and struggling for a >> little bit, and ultimately I don't think it's necessarily horrible if >> you have to be aware of this wrinkle in the Ruby environment Shoes >> creates. But it certainly would be better if we could at least help >> avoid people being caught up on it like we were. >> > > Hmm.. I don't know what other namespace issues you're referring to that this > patch would affect. Could you explain further?
Well, the Widget.inherited patch only allows for class Neato < Shoes::Widget to work, right? I raised the point that it might be nice if we had something like Rails that uses Module#const_missing and such to catch technically invalid constant references and turn them into valid ones so that we could do something like class Neato < Widget in the context of an external file. But I also said that I'm not sure going fully down this road is a good idea. I dunno, I'm tired.. maybe I'll have something more useful to say later. :) -- Seth Thomas Rasmussen http://greatseth.com
