in reply to Trying to re-use code at different levels of an inherited object.

I am no OO wizard and i really don't fully understand what you are trying to accomplish. I am intriqued by the question however, and i toyed around trying to find a solution. From what i can gather, you are trying to call SUPER after you have already 'backed up' to the SUPER object - not a recommended thing to do.

Instead, how about a more traditional polymorphic approach: 

package Foo;

use strict;

sub new {
    bless {}, shift();
}

sub do_something {
   my $self = shift;
   my $name = shift || 'foo';
   return make_do_something($name);
}

sub make_do_something {
    my $name = shift;
    return sub {
        my $self = shift;
        print "Foo $name!!!\n";
    }
}

1;

##############################################

package Bar;

use strict;
use base 'Foo';

sub new {
    bless {}, shift();
}

sub do_something {
   my $self = shift;
   return $self->SUPER::do_something('bar');
}

1;

[download]
I _think_ this will accomodate your needs ...

This is my test script. The big difference is that the do_something() method returns a code ref, so you need to add some syntax: 

#!/usr/bin/perl -w

use strict;
use Foo;
use Bar;

$_->do_something()->() for (new Foo, new Bar);

[download]
The reason why is, well - i don't know exactly what you are trying to accomplish by returning a sub in your make_do_something() method. If the only reason you do that is to create a method, then you can use this code and just drop make_do_something() completely. Everything you need is in Foo::do_something(). Hope this helps.

jeffa

L-LL-L--L-LL-L--L-LL-L--
-R--R-RR-R--R-RR-R--R-RR
B--B--B--B--B--B--B--B--
H---H---H---H---H---H---
(the triplet paradiddle with high-hat)

Replies are listed 'Best First'.
Re: (jeffa) Re: Trying to re-use code at different levels of an inherited object.
by ehdonhon (Curate) on Jan 29, 2002 at 09:19 UTC

    If it were that simple, I'd be set :)

    I'm sorry that I didn't make sense the first time. The issue is that I want to recursively call SUPER::do_something all the way up to the top most superclass that can do_something. I could do that like this:

    package A;
sub do_something {
  print "Foo\n";
}

package B;
@ISA = ( A );
sub do_something {
  my $self = shift;
  $self->SUPER::do_something();
  print "Bar\n";
}

    [download]

    But, I'd end up with a whole bunch of stub modules sitting around that were all almost identical, save a literal constant ("Foo\n" vs. "Bar\n") and the fact that A does not call SUPER::do_something(). Shouldn't I be able to abstract this all away so that I only have to implement do_something() once and not worry about making stubs in all my modules? As I said, I'm not very experienced with closures, but I thought this is what they were all about. It would be more maintainable if I could change the requirements of do_something() in the future and NOT have to go change every one of my stubs.

    I guess I should just suck it up and write the stubs.

[reply]
[d/l]

In Section Seekers of Perl Wisdom