Unexpected result when subtracting floats

Question

Possible Duplicate:
ruby floating point errors

In Ruby, when subtracting 4.7 from 5.0, I would expect a result of 0.3 but get:

~ $ irb
1.9.2p290 :001 > 5.0 - 4.7
 => 0.2999999999999998

I'm guessing there is a reason for this rather than it being a bug? Using BigDecimal objects yields the same result. Is my only option to use round on the result?

Ah, using BigDecimal with a string as the parameter gets me what I want: `(BigDecimal.new('5.0') - BigDecimal('4.7')).to_f`, but not `(BigDecimal.new(5.0, 0) - BigDecimal(4.7, 0)).to_f` — james246, Dec 04 '12 at 15:43

score 7 · Accepted Answer · edited Jan 21 '13 at 17:45

7

Floats lose precision. Nothing can be done about that. So, use:

(5.0 - 4.7).round(1)

and if you had needed more precision:

(0.50 - 0.47).round(2)

edited Jan 21 '13 at 17:45

xpda

15,585
8
51
82

answered Dec 04 '12 at 15:52

Candide

30,469
8
53
60

Figured I'd need to do this, thanks for the confirmation! – james246 Dec 04 '12 at 15:57

score 2 · Answer 2 · answered Dec 04 '12 at 15:53

2

It's not a bug. It's how floating point arithmetics works. More details here: http://floating-point-gui.de/

answered Dec 04 '12 at 15:53

DNNX

6,215
2
27
33

Thanks for the link, I knew it was _something_ to do with how Float works, but not exactly why. Will give it a read. – james246 Dec 04 '12 at 15:56

Unexpected result when subtracting floats

2 Answers2