What does it do?
Adds, multiplies the currency amounts, and calculates percentages of amounts. The result of
each of those operations is also an amount: a string, strictly matching the /^\-?\d+\.\d\d$/
pattern, like "0.25", "1000.00", or "-42.10".
Amounts on input and output are arbitrary large and precise:
99999999999999999999999999999999999999999999999999999999999999999999999999999999.99
+
0.01
=
100000000000000000000000000000000000000000000000000000000000000000000000000000000.00
However, in cases when the division is involved โ like for percentage calculation โ the result is rounded to the whole cent.
money.percent("0.50", "33.00") // is "0.17" instead of "0.165"
As a bonus feature, there's a simple formatting function for amounts in the following currencies:
- CHF
- EUR
- GBP
- JPY
- LTL
- PLN
- SEK
- SKK
- UAH
- USD
money.format("EUR", "-1560.00") // "-1.560,00"
Why does it exist?
Because storing currency amounts in floats is a really bad idea
Install
Works both on Node and in the browser.
Node/Browserify/Webpack
$ npm install --save money-math
var money = require("money-math");
Browser global
Download jsbn/index.js
Download money.js
window.Money
Usage
money.add("16.11", "17.07"); // "33.18"
money.subtract("16.00", "7.00"); // "9.00"
money.mul("24.00", "0.25"); // "6.00"
money.div("64.00", "2.00"); // "32.00"
money.percent("200.00", "3.25"); // "6.50"
money.cmp("100.00", "200.00"); // -1
money.isEqual("100.00", "100.00"); // true
money.isZero("0.00"); // true
money.isNegative("-1.00"); // true
money.isPositive("-1.00"); // false
money.isGreaterThan("2.00", "1.00"); // true
money.isGreaterOrEqualTo("2.00", "2.00"); // true
money.isLessThan("2.00", "1.00"); // false
money.isLessrOrEqualTo("2.00", "2.00"); // true
money.format("JPY", "236800.00"); // "236,800"
money.floatToAmount(56.345); // "56.35"
And last, but not least :)
money.roundUpTo5Cents("42.02"); // "42.05"
money.roundTo5Cents("442.26"); // "442.25"
Which we use for bills in CHF that are required by law to be 0 (mod 5).
An important note on the amount "data type"
The amount strings are expected to strictly adhere to the format described by the regular expression noted above. Thus, for example, it must be:
"10.10"
, not"10.1"
, not"10.100"
;"10.00"
, not10
, not"10"
, not"10.0"
.
That's a precondition for any of the API functions accepting amount arguments to work correctly. I understand that it may be confusing to some of new users; but I believe that's an optimally pragmatic way to mimic, by convention, an algebraic data type in idiomatic JavaScript -- a (very) dynamically typed language.
Just for the sake of convenience, we provide a way to approximate an imprecise float value in
the amounts field with money.floatToAmount(โฆ)
(half up rounding applied). Once all the values
are amounts, money-math guarantees that all the field operations keep the results withing the
field. Classic algebra.
A thoughtful reader might ask, why have money.floatToAmount()
, when there's the
Number.prototype.toFixed(2)
? Well, because:
> 56.155.toFixed(2);
'56.16'
> 56.345.toFixed(2);
'56.34'
Floats are such floats...