Because they're floating point numbers - so are always approximations:
irb(main):006:0> sprintf('%.20f', 0.565)
=> "0.56499999999999994671"
irb(main):007:0> sprintf('%.20f', 0.555)
=> "0.55500000000000004885"
On Dec 14, 2009, at 2:43 PM, William Cassis wrote:
> Hi folks,
>
> Can anyone help explain the following Ruby (specifically ruby
> 1.9.1p243
> (2009-07-16 revision 24175)) Float rounding behavior?:
>
>> irb(main):001:0> 0.555.round(2)
>> => 0.56
>
> Looks correct
>
>> irb(main):002:0> 0.565.round(2)
>> => 0.56
>
> Why?
>
>> irb(main):003:0> 0.575.round(2)
>> => 0.57
>
> Why?
>
>> irb(main):004:0> 0.585.round(2)
>> => 0.59
>
> Looks correct
>
> If I instead convert the decimals above into BigDecimal objects and
> then
> perform the rounding operation it produces the "expected" output
> everytime.
>
> The rounding example below happens on both Ruby 1.9 and 1.8.6 (ruby
> 1.8.6 (2008-08-11 patchlevel 287)):
>
>> irb(main):001:0> 57.5.class
>> => Float
>> irb(main):002:0> 57.5.round
>> => 58
>
> Looks correct
>
>> irb(main):003:0> (Float(0.575) * (10 ** 2))
>> => 57.5
>> irb(main):004:0> (Float(0.575) * (10 ** 2)).class
>> => Float
>> irb(main):005:0> (Float(0.575) * (10 ** 2)).round
>> => 57
>
> Why?
>
> Thanks for your help!
>
> --
> SD Ruby mailing list
> [email protected]
> http://groups.google.com/group/sdruby
--
SD Ruby mailing list
[email protected]
http://groups.google.com/group/sdruby
