From: "Michael W. Ryder" <_mwryder@...>
Date: 2008-05-22T04:10:06+09:00
Subject: Re: Why doesn't Float() work the same as Integer()?

Yukihiro Matsumoto wrote:
> Hi,
> 
> In message "Re: Why doesn't Float() work the same as Integer()?"
>     on Tue, 20 May 2008 11:30:12 +0900, "Eric I." <rubytraining@gmail.com> writes:
> 
> |First, Float() and Integer() work like Ruby itself.  In Ruby code you
> |can hardcode integer literals using 17 (dec), 021 (oct), or 0x11
> |(hex).  But you can only hardcode floating point literals using base
> |ten.  Admittedly, that's a bit of a punt since you'll probably ask why
> |Ruby works that way.
> 
> Because most of us would have hard time to guess how much 0x11.4 is.
> Besides that, I don't see any usecase for converting hexadecimal
> (integer) representation directly into float values.
> 
> 							matz.
> 
Thank you for the reply.  I wanted to convert strings to Floats first as 
it simplified some of a method's logic.  The method allows 
Rational.new(3.5) and returns 7/2, it also allows Rational.new(3.5, 2) 
and returns 7/4.  Converting from strings seemed to be a simple addition 
to the method and rather than having separate logic for Integers and 
Floats I wanted to just convert all Strings to Floats and then convert 
the Floats.  Since I had already gone this far I decided to add support 
for any number that Ruby allows, this is where I ran into the 
"inconsistency" in the methods Float(x) and Integer(x).
As Ruby doesn't appear to support Floats in any base other than 10 I had 
no problem if it didn't convert '0x11.4' just like it wouldn't convert 
'abcd'.  But since it could convert valid Integers I was confused while 
it couldn't convert '0x11'.