Re: semantics of CMOP's load_class()

2009-09-22 Thread Jesse Luehrs
On Tue, Sep 22, 2009 at 08:43:08PM -0500, Jesse Luehrs wrote: > On Tue, Sep 22, 2009 at 06:16:48PM -0700, Darren Duncan wrote: > > 1. I looked for the implementation of is_class_loaded() and it seems the > > only implementation is XS, in the file MOP.xs, with no Pure Perl version > > existing, a

Re: semantics of CMOP's load_class()

2009-09-22 Thread Darren Duncan
As an addendum, I suppose part of my project design is antiquated and I could simply ditch the ::Interface-provided new_machine() wrapper function entirely, in which case all implementation code would just be in the implementation, like all the other wrappers. So then users just say something l

Re: semantics of CMOP's load_class()

2009-09-22 Thread Jesse Luehrs
On Tue, Sep 22, 2009 at 06:16:48PM -0700, Darren Duncan wrote: > Now I recognize that Class::MOP provides these, and other, utility functions: > > Class::MOP::is_class_loaded($class_name) > Class::MOP::load_class($class_name) > > ... and that these ostensibly are better used instead of doing it

semantics of CMOP's load_class()

2009-09-22 Thread Darren Duncan
Hello, So I have some Perl code which has evolved over the years, which already uses Moose, but I'm looking to make it more savvy or modern/enlightened. And I've also found what appear to be bugs or deficiencies in Class::MOP. The original code that I'm concerned about updating is a single su