NAME
   Algorithm::C3 - A module for merging hierarchies using the C3 algorithm

SYNOPSIS
     use Algorithm::C3;

     # merging a classic diamond
     # inheritance graph like this:
     #
     #    <A>
     #   /   \
     # <B>   <C>
     #   \   /
     #    <D>

     my @merged = Algorithm::C3::merge(
         'D',
         sub {
             # extract the ISA array
             # from the package
             no strict 'refs';
             @{$_[0] . '::ISA'};
         }
     );

     print join ", " => @merged; # prints D, B, C, A

DESCRIPTION
   This module implements the C3 algorithm. I have broken this out into
   it's own module because I found myself copying and pasting it way too
   often for various needs. Most of the uses I have for C3 revolve around
   class building and metamodels, but it could also be used for things like
   dependency resolution as well since it tends to do such a nice job of
   preserving local precedence orderings.

   Below is a brief explanation of C3 taken from the Class::C3 module. For
   more detailed information, see the "SEE ALSO" section and the links
   there.

 What is C3?
   C3 is the name of an algorithm which aims to provide a sane method
   resolution order under multiple inheritance. It was first introduced in
   the language Dylan (see links in the "SEE ALSO" section), and then later
   adopted as the preferred MRO (Method Resolution Order) for the new-style
   classes in Python 2.3. Most recently it has been adopted as the
   'canonical' MRO for Perl 6 classes, and the default MRO for Parrot
   objects as well.

 How does C3 work.
   C3 works by always preserving local precedence ordering. This
   essentially means that no class will appear before any of it's
   subclasses. Take the classic diamond inheritance pattern for instance:

        <A>
       /   \
     <B>   <C>
       \   /
        <D>

   The standard Perl 5 MRO would be (D, B, A, C). The result being that A
   appears before C, even though C is the subclass of A. The C3 MRO
   algorithm however, produces the following MRO (D, B, C, A), which does
   not have this same issue.

   This example is fairly trivial, for more complex examples and a deeper
   explanation, see the links in the "SEE ALSO" section.

FUNCTION
   merge ($root, $func_to_fetch_parent, $cache)
       This takes a $root node, which can be anything really it is up to
       you. Then it takes a $func_to_fetch_parent which can be either a
       CODE reference (see SYNOPSIS above for an example), or a string
       containing a method name to be called on all the items being
       linearized. An example of how this might look is below:

         {
             package A;

             sub supers {
                 no strict 'refs';
                 @{$_[0] . '::ISA'};
             }

             package C;
             our @ISA = ('A');
             package B;
             our @ISA = ('A');
             package D;
             our @ISA = ('B', 'C');
         }

         print join ", " => Algorithm::C3::merge('D', 'supers');

       The purpose of $func_to_fetch_parent is to provide a way for "merge"
       to extract the parents of $root. This is needed for C3 to be able to
       do it's work.

       The $cache parameter is an entirely optional performance measure,
       and should not change behavior.

       If supplied, it should be a hashref that merge can use as a private
       cache between runs to speed things up. Generally speaking, if you
       will be calling merge many times on related things, and the parent
       fetching function will return constant results given the same
       arguments during all of these calls, you can and should reuse the
       same shared cache hash for all of the calls. Example:

         sub do_some_merging {
             my %merge_cache;
             my @foo_mro = Algorithm::C3::Merge('Foo', \&get_supers, \%merge_cache);
             my @bar_mro = Algorithm::C3::Merge('Bar', \&get_supers, \%merge_cache);
             my @baz_mro = Algorithm::C3::Merge('Baz', \&get_supers, \%merge_cache);
             my @quux_mro = Algorithm::C3::Merge('Quux', \&get_supers, \%merge_cache);
             # ...
         }

CODE COVERAGE
   I use Devel::Cover to test the code coverage of my tests, below is the
   Devel::Cover report on this module's test suite.

    ------------------------ ------ ------ ------ ------ ------ ------ ------
    File                       stmt   bran   cond    sub    pod   time  total
    ------------------------ ------ ------ ------ ------ ------ ------ ------
    Algorithm/C3.pm           100.0  100.0  100.0  100.0  100.0  100.0  100.0
    ------------------------ ------ ------ ------ ------ ------ ------ ------
    Total                     100.0  100.0  100.0  100.0  100.0  100.0  100.0
    ------------------------ ------ ------ ------ ------ ------ ------ ------

SEE ALSO
 The original Dylan paper
   <http://www.webcom.com/haahr/dylan/linearization-oopsla96.html>

 The prototype Perl 6 Object Model uses C3
   <http://svn.openfoundry.org/pugs/perl5/Perl6-MetaModel/>

 Parrot now uses C3
   <http://aspn.activestate.com/ASPN/Mail/Message/perl6-internals/2746631>
   <http://use.perl.org/~autrijus/journal/25768>

 Python 2.3 MRO related links
   <http://www.python.org/2.3/mro.html>
   <http://www.python.org/2.2.2/descrintro.html#mro>

 C3 for TinyCLOS
   <http://www.call-with-current-continuation.org/eggs/c3.html>

AUTHORS
   Stevan Little, <[email protected]>

   Brandon L. Black, <[email protected]>

COPYRIGHT AND LICENSE
   Copyright 2006 by Infinity Interactive, Inc.

   <http://www.iinteractive.com>

   This library is free software; you can redistribute it and/or modify it
   under the same terms as Perl itself.

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.