From: "Randy W. Sims" <ml-ruby@...>
Date: 2004-10-09T18:18:12+09:00
Subject: Re: Range behavior (Re: [RCR] New [] Semantics)

On 10/9/2004 5:08 AM, Yukihiro Matsumoto wrote:
> Hi,
> 
> In message "Re: Range behavior (Re: [RCR] New [] Semantics)"
>     on Sat, 9 Oct 2004 17:49:57 +0900, "Randy W. Sims" <ml-ruby@thepierianspring.org> writes:
> 
> |The behavior of:
> |
> |irb(main):002:0> (1..5).include?( 2.5 )
> |=> true
> |
> |was a complete suprise to me. It goes against the expectations created 
> |by the first example above.
> 
> The point is whether your expectation is
> 
>   * important enough to keep, even after you've been explained
>   * worthy for the cost of iteration, such that
> 
>       (0..1000000000).member?(999999999)
> 
>     takes seven minutes.

Hmm, I guess I expected Range to be implemented as an ordered set, an 
array. Thus Range#member? would be a quick binary search...

Randy.