On NullClass or FalseClass#method_missing

Hi--

George Mischovitis' Nitro project uses OprnStruct-type containers
extensively for their end-user ease of use. I have been working with
him to make the solution more robust. ONe difficulty we've run into is
avoiding an error on an unregonized string of calls:

  o = OpenStructLikeClass
  o.x     #=> nil
  o.x.y  #=> NoMethodError

George needs:

  o = OpenStructLikeClass
  o.x     #=> nil
  o.x.y  #=> nil

I corrected by creating a new class like NilClass, called NullClass:

  class NullClass #< NilClass
    class << self
      def new
        @null ||= NullClass.allocate
      end
    end
    def inspect ; 'null' ; end
    def nil?  ; true ; end
    def null? ; true ; end
    def method_missing(*args) self ; end
  end

  module Kernel
    def null
      NullClass.new
    end
  end

  class Object
    def null?
      false
    end
  end

As you can see I wanted to subclass NilClass, but this proved
impossible, b/c there is no way to instantiate the class. Becasue of
this one can't do

  unless o.x.y  #null

because null is considered true. So that's problematic to George too.

So I'm wondering if Ruby might want to take this into account for the
future, either by adding NullClass, or allowing for subclasses of
NilClass.

The only alternative I've come up with --just recently, is to put the
mehtod_missing in FalseClass. I don't think anyone ever depends on
falase not responding to a method call. Does anyone see a problem with
that approach. Is it reasonable?

\ o)  カラチ
 //    trans.
/ \    [e-mail]