JavaScript types: numeric types

Representation

All numbers are represented as 64-bit IEEEE 754 floating-point values; there is no integer type. Integers between -2^53 and 2^53 can be represented exactly.

Array indexing and some bitwise operations may be performed with 32-bit integers.

Literals

Hexadecimal literals are written with the 0x or 0X prefix. Some implementations may support octal literals written with the 0 prefix, but this is not supported by the ECMAScript standard, and ECMAScript 5 strict mode forbids octal literals. As usual, you should never write an integer literal with a leading zero unless you want an octal literal, and in the case of JavaScript, since they are unsupported or forbidden, you shouldn’t use them at all.

Floating-point literals can be written in real number syntax (using a decimal point) or exponential notation. JavaScript has no decimal numeric type like Java’s BigDecimal class or Python’s decimal module.

Arithmetic

Arithmetic operations are provided by = - * / % and by the methods of the Math object.

No errors are raised in cases of overflow, underflow, or division by zero. Overflow returns the global variable Infinity or -Infinity; underflow returns 0 or -0; division by zero returns Infinity or -Infinity, except in case of zero divided by zero, which returns the global variable NaN.

Division of Infinity by Infinity, taking the square root of a negative number, or arithmetic with non-coerceable non-numeric operands will also return NaN.

More on `NaN` and `-0`

NaN does not compare equal to any other value, including itself: NaN === NaN will return false. For this reason, you cannot use an expression like x == NaN. Instead, use isNaN(). A companion method is isFinite(), which will return true for anything but NaN, Infinity, or -Infinity.

-0 compares equal to positive zero, even using strict equality testing: -0 === 0 will return true.

Dates

Date handling is provided by methods on the core object Date.

References