Jonathan Rockway wrote:
> * On Wed, Apr 08 2009, Michael G Schwern wrote:
>> # Moose
>> sub DOES {
>> return $_[0]->meta->does_role($_[1]);
>> }
>>
>> # Class::Trait
>> sub DOES {
>> return $_[0]->does($_[1]);
>> }
>
> Why is the Perl 5.10 way ALL CAPS, anyway? DOES is not automatically
> called by Perl, is it? (Follow-up question: why is isa not ISA?)
I'd guess the thinking was it's less likely to clash with an existing does()
method for those 3 programs out there which will freak out if
$obj->can("does") suddenly starts being true.
I'm psychic.
http://groups.google.com/group/perl.perl5.porters/msg/c75227ea651e0c7e
isa() is not ISA() because it was added before Perl 5 OO was widespread enough
so as to be so paranoid about new UNIVERSAL methods being added. That being
around 5.003.
Interestingly enough, when UNIVERSAL came in there were a few extra methods,
class() and is_instance().
http://perl5.git.perl.org/perl.git/commit/6d4a7be2b18d1674acf2ccc0da715a204e2d1ed0
They were junked for being redundant. Oddly enough, is_instance() would
return as Scalar::Util::blessed() except it works on non-objects which was one
of the objections to is_instance() and continues to nag us with unblessed refs.
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/1997-02/msg00683.html
Amusingly, someone proposes this mythical "perl6" thing that might break the
"$class = ref $obj" idiom.
http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/1997-02/msg00691.html
--
If you want the truth to stand clear before you, never be for or against.
The struggle between "for" and "against" is the mind's worst disease.
-- Sent-ts'an
