From: "David A. Black" <dblack@...>
Date: 2001-09-04T02:51:58+09:00
Subject: [ruby-talk:20796] Re: RCR: non-bang equivalent to []=

Hello --

On Tue, 4 Sep 2001, Kevin Smith wrote:

> --- ts <decoux@moulon.inra.fr> wrote:
> > >>>>> "K" == Kevin Smith <kevinbsmith@yahoo.com> writes:
> > 
> > K> class Array
> > K>   def copy
> > K>     return []+self
> > K>   end
> > K> end
> > 
> >  Probably I'm missing something but you have written
> #clone
> 
> No, you're not missing anything. I think this would be a
> handy method to include in the core Array class. I think
> the name copy is better than the name clone, because clone
> has special meanings to people.
> 
> It's not a killer, but it took me maybe twenty minutes to
> figure out this simple way to copy an array. It seems like
> a useful thing to do, and it's something I have wanted to
> do in the past.
> 
> Also, it could be improved to "fix" the problem you
> mentioned earlier. If a contains itself, then a copy of a
> should contain a copy of _itself_, and not of a.
> Immediately after doing []+self, it should be easy to
> replace all elements that are 'self' with references to the
> newly-created array.
> 
> Does anyone else think this is worth of RCR status?


Are you suggesting changing #clone's name to #copy?


David

-- 
David Alan Black
usually dblack@candle.superlink.net
writing from dblack@saturn.superlink.net