From: Ben Tilly <ben_tilly@...>
Date: 2001-02-08T22:17:10+09:00
Subject: [ruby-talk:10571] Re: Giving a Proc utility methods?

wys@helbling.ch (Clemens Wyss) wrote:
>
>Aleksi Niemel<zak@ale.cx> wrote in
><Pine.LNX.4.21.0102081209130.1404-100000@otake>:
> >I meant it seems the proc has hard times to refer to itSelf at
> >runtime. I'd guess this is actually pretty common operation (although
> >I haven't used it ever) because it's pretty common for normal methods to
> >refer the instance object with self.
> >
>...again, "self" in the block-code is NOT the Proc object but the "main
>context" of the program. And "def meth.fail(msg)" does not extend (the
>context of) "self" but the Proc object. Your solution does look closer to
>Ben's but object-wise it isn't.

Both of your solutions work.  Personally I prefer passing
in the current object explicitly because then I don't need
to do so much work if I want to bind more methods, plus
it is then easier to subclass my class with even more
utility methods.

I still wish there was a way to make the original idea
work though.  But I guess I can live without.

>I don't see real need for accessing the Proc object.

This is the second time I have tried to access the Proc
object and failed to get it working correctly.  The
other time I was trying to create an anonymous recursive
subroutine.  (Just to see if I could.)  Oops.

So accessing the Proc object from within the anonymous
subroutine is something that I could see uses for.  But
I can't say that the feature is worth the complexity.

Cheers,
Ben
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com