How to prevent Ruby money floating point errors

Question

I'm using Rails with the money-rails gem to handle money columns.

Is there any way to prevent floating point errors from occuring? (even a hack will do, I just want to make sure no such errors are presented to the end user)

Rspec example case:

  it "correctly manipulates simple money calculations" do
    # Money.infinite_precision = false or true i've tried both 
    start_val = Money.new("1000", "EUR")
    expect(start_val / 30 * 30).to eq start_val
  end

Result

Failure/Error: expect(start_val / 30 * 30).to eq start_val

   expected: #<Money fractional:1000.0 currency:EUR>
        got: #<Money fractional:999.99999999999999999 currency:EUR>

   (compared using ==)

   Diff:
   @@ -1,2 +1,2 @@
   -#<Money fractional:1000.0 currency:EUR>
   +#<Money fractional:999.99999999999999999 currency:EUR>

Danil Gaponov · Accepted Answer · 2015-08-25T11:42:37.693

6

You should use decimals for money amounts. See http://ruby-doc.org/stdlib-2.1.1/libdoc/bigdecimal/rdoc/BigDecimal.html for instance. It has arbitrary precision arithmetic.

EDIT: In your case you should probably change your Rspec to something like:

it "correctly manipulates simple money calculations" do
  # Money.infinite_precision = false or true i've tried both 
  start_val = Money.new("1000", "EUR")
  thirty = BigDecimal.new("30")
  expect(start_val / thirty * thirty).to eq start_val
end

EDIT2: in this very case 1000/30 cannot be represented as a finite decimal number. You have to either use Rational class or do rounding. Example code:

it "correctly manipulates simple money calculations" do
  # Money.infinite_precision = false or true i've tried both 
  start_val = Money.new("1000", "EUR")
  expect(start_val.amount.to_r / 30.to_r * 30.to_r).to eq start_val.amount.to_r
end

edited Aug 25 '15 at 11:42

answered Aug 25 '15 at 10:49

Danil Gaponov

1,413
13
23

1

If you take a look at the Ruby money gem source, ([link](https://github.com/RubyMoney/money/blob/master/lib/money/money.rb#L251)), it is already handling the amounts as BigDecimals – Spyros Mandekis Aug 25 '15 at 10:59
no use, plus : `1000.to_d / 30.to_d * 30.to_d #=> #` – Spyros Mandekis Aug 25 '15 at 11:09
3

1000/30 cannot be finitely represented in decimal. Period. It doesn't really matter what decimal data type you use, unless you use a computer with infinite memory, 1000/30 *must* be rounded. It *can* be represented in ternary, though: 1020.1. – Jörg W Mittag Aug 25 '15 at 11:30
Yep, I see, he has to use Rationals if they exist in ruby. – Danil Gaponov Aug 25 '15 at 11:39
The `Rational` class was exactly what I was looking for, thanks :) – Spyros Mandekis Aug 25 '15 at 11:46

How to prevent Ruby money floating point errors

1 Answers1

Linked