From: Austin Ziegler <austin@...>
Date: 2004-03-02T04:05:53+09:00
Subject: Re: concerns about Proc,lambda,block

On Tue, 2 Mar 2004 03:23:41 +0900, David A. Black wrote:
> Hi --
>
>
> On Tue, 2 Mar 2004, ts wrote:
>
>
>>>>>>> "D" == David A Black <dblack@wobblini.net> writes:
>>>>>>>
>>
>> D> I think Dave's explanation is about as good as one can give -- and
>> D> that's what worries me.  In particular, the fact that Ruby is now D>
>> doing something that requires the concept of "contexts" to explain D>
>> strikes me as an indication of a possible problem.
>>
>> Well, the problem is perhaps that he try to rationalize (he must be
>> cartesian :-)) when you must "visually" see the examples.
>>
>
> But there isn't any visual information to go on, once these objects leave
> the site of origin.  Nor do the objects differentiate themselves very
> clearly:
>
> irb(main):012:0> n.methods - m.methods
> => []
> irb(main):013:0> n.class == m.class
> => true
> irb(main):014:0> n.arity == m.arity
> => true
> irb(main):015:0> n.call
> ArgumentError: wrong number of arguments (0 for 2) from (irb):15:in `x'
> from (irb):8:in `call' from (irb):15 irb(main):016:0> m.call => nil
> irb(main):017:0> RUBY_VERSION => "1.9.0"
>
> There's nothing wrong with having code blocks/closures that behave
> differently.  But it isn't represented in the language the way that other
> such differences between built-ins are.  (And the actual rules having to
> do with arity also seem very complex to me.)

I think that what's more bothersome to me is that there doesn't appear to be 
a way to convert one type of Proc to another. I can't do: 
n.argument_style(:yield) or n.argument_style(:method).

-austin
--
austin ziegler    * austin@halostatue.ca * Toronto, ON, Canada
software designer * pragmatic programmer * 2004.03.01
                                         * 14.02.40