[ruby-core:17376] Re: Any reason for having no module exclusion functionality in Ruby

I want to +1 this again and reraise it for consideration.
Matz, think about this like the poor-man's selector namespaces. It wouldn't
be as robust as real namespaces, but it would allow the injection of
functionality for a small period of time, and then uninject the
functionality.

Specifically, this can be used in frameworks to pull in modules that can be
potentially polluting (e.g. a module of Helpers for a View class) and then
remove the pollution once the needed functionality has been exercised.

In Merb, in particular, it would be nice to be sure that helpers didn't
override built-in controller methods. At the moment, we do a fairly ugly
thing of naming internal methods with _method, so that it's less likely for
the pollution to do anything dangerous, but a much more elegant solution
would be: include Helpers; result =3D render; uninclude Helpers;
finish_dispatch.

-- Yehuda

On Wed, Apr 23, 2008 at 10:22 PM, Kurt Stephens <[e-mail]> wrote=
:

> Yehuda Katz wrote:
> > +1.
> >
> > Being able to unextend modules would make it possible to temporarily
> > give methods to an object that were easily removable. I've had a number
> > of use-cases for it in the past few months.
> >
> > -- Yehuda
>
> +1
>
> We do something similar when testing legacy code that is difficult to
> decouple.  We have an DSL object that can redefine methods before a
> block, yields to a block and then restores the methods after the block
> completes.
>
> It would be so much cleaner if we could just extended and unextend.  It
> would also be nice to do the same with include: uninclude?
>
> -- Kurt Stephens
>
> >
> > On Apr 23, 2008, at 3:50 PM, Pit Capitain wrote:
> >
> >> Hi all, I'm forwarding the following message for Yurii, who seems to
> >> have problems joining ruby-core. The current implementation can be
> >> found here:
> >>
> >> http://github.com/yrashk/rbmodexcl/tree/master
> >>
> >> Regards,
> >> Pit
> >>
> >> ---------- Forwarded message ----------
> >> From: Yurii Rashkovskii <[e-mail]>
> >> Date: 2008/4/22
> >> Subject: Any reason for having no module exclusion functionality in R
> >> To: ruby-talk ML <[e-mail]>
> >>
> >>
> >> Hello,
> >>
> >> May be I am missing something (since I am not MRI hacker), but why
> >> there is no module exclusion functionality =97 i.e. you can extend
> >> object with module, but can't reverse this operation? Any specific
> >> reason for this decision?
> >>
> >> I've just sketched simplistic implementation for this kind of
> >> functionality (http://pastie.org/185000) and really wondering why
> >> there is no such stuff in MRI.
> >>
> >> Again, am I missing something?
> >>
> >> Yurii.
> >> --
> >> Posted via http://www.ruby-forum.com/.
> >>
> >
> >
>
>
>


--=20
Yehuda Katz
Developer | Engine Yard
(ph) 718.877.1325