Math, Numbers & Tez

LIGO offers three built-in numerical types: int, nat and tez. Values of type int are integers; values of type nat are natural numbers (integral numbers greater than or equal to zero); values of type tez are units of measure of Tezos tokens.

• Integer literals are the same found in mainstream programming languages, for example, 10, -6 and 0, but there is only one canonical zero: 0 (so, for instance, -0 and 00 are invalid).

• Natural numbers are written as digits followed by the suffix n, like so: 12n, 0n, and the same restriction on zero as integers applies: 0n is the only way to specify the natural zero.

• Tezos tokens can be specified using literals of three kinds:

  • units of millionth of tez, using the suffix mutez after a natural literal, like 10000mutez or 0mutez;
  • units of tez, using the suffix tz or tez, like 3tz or 3tez;
  • decimal amounts of tz or tez, like 12.3tz or 12.4tez.

Note that large integral values can be expressed using underscores to separate groups of digits, like 1_000mutez or 0.000_004tez.

Addition

Addition in LIGO is accomplished by means of the + infix operator. Some type constraints apply, for example you cannot add a value of type tez to a value of type nat.

In the following example you can find a series of arithmetic operations, including various numerical types. However, some bits remain in comments as they would otherwise not compile, for example, adding a value of type int to a value of type tez is invalid. Note that adding an integer to a natural number produces an integer.

// int + int yields int
const a = 5 + 10;

// nat + int yields int
const b = 5n + 10;

// tez + tez yields tez
const c: tez = 5mutez + 1tez;

// tez + int or tez + nat is invalid:
// const d : tez = 5mutez + 10n;

// two nats yield a nat
const e: nat = 5n + 10n;

// nat + int yields an int: invalid
// const f : nat = 5n + 10;

const g = 1_000_000;

Tip: you can use underscores for readability when defining large numbers:

let sum : tez = 100_000mutez;

Subtraction

Subtraction looks as follows.

⚠️ Even when subtracting two nats, the result is an int.

const a = 5 - 10;

// Subtraction of two nats yields an int
const b: int = 5n - 2n;

// Therefore the following is invalid
// const c : nat = 5n - 2n;

From protocol Ithaca onwards subtracting values of type tez yeilds an optional value (due to the Michelson instruction SUB_MUTEZ)

const d : option<tez> = 5mutez - 1mutez; /* Some (4mutez) */
const e : option<tez> = 1mutez - 5mutez; /* None */

Multiplication

You can multiply values of the same type, such as:

const a = 5 * 5;
const b: nat = 5n * 5n;

// You can also multiply `nat` and `tez`
const c: tez = 5n * 5mutez;

Euclidean Division

In LIGO you can divide int, nat, and tez. Here is how:

⚠️ Division of two tez values results into a nat.

const a: int = 10 / 3;
const b: nat = 10n / 3n;
const c: nat = 10mutez / 3mutez;

LIGO also allows you to compute the remainder of the Euclidean division. In LIGO, it is a natural number.

The behaviour of the % operator in JsLIGO is different from JavaScript. In JsLIGO, % is a modulus operator and in JavaScript it's a remainder operator. In the case of positive numbers everything is the same, but not with negative numbers.

const a = 120;
const b = 9;
const rem1 = a % b;  // 3
const c = 120n;
const rem2 = c % b;  // 3
const d = 9n;
const rem3 = c % d;  // 3
const rem4 = a % d;  // 3

For cases when you need both the quotient and the remainder, LIGO provides the ediv operation. ediv x y returns Some (quotient, remainder), unless y is zero, in which case it returns None

const a = 37;
const b = 5;
const ediv1 : option<[int , nat]> = ediv(a, b);  // Some (7, 2)
const c = 37n;
const ediv2: option<[int , nat]> = ediv(c, b);  // Some (7, 2)
const d = 5n;
const ediv3: option<[nat , nat]> = ediv(c, d);  // Some (7, 2)
const ediv4: option<[int , nat]> = ediv(a, d);  // Some (7, 2)

From int to nat and back

You can cast an int to a nat and vice versa. Here is how:

const a = int(1n);
const b = abs(1);

Checking a nat

You can check if a value is a nat by using a predefined cast function which accepts an int and returns an optional nat: if the result is not None, then the provided integer was indeed a natural number, and not otherwise.

const is_a_nat = is_nat(1);

Increment operator

Increment opeator increments (adds one to) the value of the binder.

In the prefix position (++p) the operator increments the value and returns the latest incremented value.

In the postfix position (p++) the operator increments the value but returns the old value before the increment.

const testInc = (() => {
  let inc = 0;

  // Prefix increment operator
  assert(++inc == 1);
  assert(inc   == 1);

  // Postfix increment operator
  assert(inc++ == 1);
  assert(inc   == 2);
})();

Decrement operator

Decrement opeator decrements (subtracts one from) the value of the binder.

In the prefix position (--p) the operator decrements the value and returns the latest decremented value.

In the postfix position (p--) the operator decrements the value but returns the old value before the decrement.

const testDec = (() => {
  let v = 10;

  // Prefix decrement operator
  assert(--v == 9);
  assert(v   == 9);

  // Postfix decrement operator
  assert(v-- == 9);
  assert(v   == 8);
})();