From: Sean Russell <ser@...>
Date: 2001-11-07T22:21:24+09:00
Subject: [ruby-talk:24507] Range weirdness

Hi all,

ruby 1.6.2 (2001-01-23) [i586-linux-gnu]

I've got a class that I'd like to be a number, so I extended Numeric.  I 
discovered, however, that when you try to turn these sorts of classes into 
Ranges, you get some odd, yet predictable, behavior.  

Here's an illustration:

  class RomanNumerals < Numeric
    ...
  end

  x = (RomanNumerals.new(1)..RomanNumerals.new(10))
  # -> (I..X)

so get a proper range.  However, for some reason I've yet to fathom, 
Range.each calls to_i on any Numerics it finds in the range, and goes from 
there.  Therefore, if you've defined to_i for RomanNumerals, you get:

  x.to_a
  # -> [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]

Behold the Holy Stack Trace:

  c-call -e:1        to_a Enumerable
  c-call -e:1        each    Range
    call ./ternary.rb:56       to_i BTernary
    line ./ternary.rb:57       to_i BTernary
  return -e:1        to_i BTernary
    call ./ternary.rb:56       to_i BTernary
    line ./ternary.rb:57       to_i BTernary
  return -e:1        to_i BTernary

Upon encountering "I", Range calls to_i, which returns 1, which it calls 
succ on, which of course returns 2, and so on.

Although I'm sure Matz has a good reason for calling to_i on all Numerals, 
this behavior is a bit unexpected.  I'd like to be able to drop in number 
replacements, but perhaps I'm subclassing the wrong class?

Thanks,

--- SER


-----=  Posted via Newsfeeds.Com, Uncensored Usenet News  =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
 Check out our new Unlimited Server. No Download or Time Limits!
-----==  Over 80,000 Newsgroups - 19 Different Servers!  ==-----