From: Pit Capitain <pit@...>
Date: 2002-02-15T00:11:58+09:00
Subject: Re: qualms about respond_to? idiom

On 14 Feb 2002, at 20:15, David Alan Black wrote:

> Hi --
> 
> On Thu, 14 Feb 2002, Avi Bryant wrote:
> 
> > > All tags respond to tagStart? and tagEnd?  But not all events in
> > > my stream are tags.  So....
> > >
> > >          if e.respond_to? :tagStart?
> > >            context.push(e.name) if e.tagStart?
> > >          end
> > >
> > > is my way of testing, first, whether it's a tag (but without
> > > actually checking whether it's a tag), and, second, whether it's a
> > > start tag. This is a case where I wish I knew a way to avoid the
> > > duplication, which feels really clunky to me.
> >
> > Ok, how about:
> >
> > class Object
> >   def try_sending(msg, *args)
> >      if respond_to? msg
> >        yield send(msg, *args)
> >      end
> >   end
> > end
> >
> > e.try_sending(:tagStart?) {|s| context.push(e.name) if s}
> 
> I certainly like the general idea of something that does this in one
> step, though I think it would be hard to handle iterators (since
> you've already got a block).  Also, it doesn't let you branch on
> non-finding of the method.  That's perhaps an insurmountable problem
> with combining these steps (and not using exceptions): how do you flag
> that a method wasn't found, as opposed to finding a method that
> returns false or nil?

Hi David,

I'd just like to remind you of ruby-talk:33770 where I showed a 
possible implementation that deals with the method-not-found 
case. Got no answer to this yet.

Pit