From: Rich Kilmer <rich@...>
Date: 2002-05-02T12:18:28+09:00
Subject: RE: Scope of Singleton Classes

See comment below:

> -----Original Message-----
> From: Jason Voegele [mailto:jason@jvoegele.com]
> Sent: Wednesday, May 01, 2002 3:03 PM
> To: ruby-talk ML
> Subject: Scope of Singleton Classes
> 
> I'm writing an object proxy that intercepts method calls with
> method_missing
> and then forwards them to another object for handling.  I want to
avoid
> the
> overhead of using method_missing every time, so I'd like to define a
new
> method each time method_missing is invoked that actually does the
> forwarding.  I have a few questions about this approach.
> 
> I tried something like this:
> 
> class ObjectProxy
>     def method_missing(operation, *args)
>         class << self
>             define_method(operation, args) do |opargs|
>                 forward(*opargs)
>             end
>         end
>     end
> 
>     def forward(operation, *args)
>         # call operation on target object
>     end
> end
> 
> 1) I was surprised to learn that the singleton class defined by "class
<<
> self" does not have access to local variables in its lexical scope
(e.g.
> "operation" and "args").  Is this by design?
> 
> 2) Is there solution that utilizes define_method rather than eval?
> 
> 3) Is caching methods like this (rather than relying on method_missing
for
> each invocation) a significant enough performance improvement to make
it
> worth it?

Because the method you define to "cache" ends up calling the #forward
method, I can't see how this would be different than performance of
calling #forward from method_missing.

One other thing to remember in your ObjectProxy is that method_missing
will not be called for methods that are defined (or included from
Modules) in Object which is quite a list.  You may want to "undef" all
the inherited methods.


> 
> Thanks,
> --
> Jason Voegele
> "We believe that we invent symbols. The truth is that they invent us."
>     -- Gene Wolfe, The Shadow of the Torturer