From: Thomas E Enebo <tom.enebo@...>
Date: 2012-04-13T23:02:14+09:00
Subject: [ruby-core:44339] Re: Bigdecimal / 1.0 => Float?

On Wed, Apr 11, 2012 at 7:17 PM, Nobuyoshi Nakada <nobu@ruby-lang.org> wrote:
> Hi,
>
> (12/04/12 0:17), Thomas E Enebo wrote:
>> b = BigDecimal("1") / 1.0
>> p b, b.class
>>
>> I was surprised to see b.class print out as a Float. �Can someone
>> explain why this is?
>
> While BigDecimal has precision info, but Float doesn't. �So `b' also
> doesn't have. �Therefore it can't be BigDecimal.
>
> Does this make sense?

yeah I wondered if it was mixing known precision with unknown.  I
guess it just looked weird because if the Bigdecimal had more
precision than a float it could contain it would lose something.  With
that said it is being divided by an inprecise type...

Thanks for the explanation,
-Tom
>
> --
> Nobu Nakada
>



-- 
blog: http://blog.enebo.com� � �� twitter: tom_enebo
mail: tom.enebo@gmail.com