JS Foundations #10 — Numbers & Math

JavaScript has one number type — IEEE 754 double-precision 64-bit floating point. That single fact explains NaN, Infinity, 0.1 + 0.2 !== 0.3, and +0/-0. This article covers Number methods, the Math object, and the traps interviewers test.

1. Number Representation

JavaScript uses 64-bit IEEE 754. Integers are safe up to 2^53 - 1; beyond that precision degrades.

For integers larger than MAX_SAFE_INTEGER, use BigInt:

2. NaN — Not a Number

NaN is the result of a failed numeric operation. It's the only value not equal to itself.

Interview use: Deep equality checks (#14), JSON.stringify (NaN → null).

3. Infinity and Finite Checks

Interview use: JSON.stringify (Infinity → null), range checks (#30).

4. +0 and -0

IEEE 754 has signed zero. +0 === -0 is true, but Object.is(+0, -0) is false.

Interview use: Deep equality — Object.is or custom check (#14).

5. Floating-Point Precision

0.1 + 0.2 is not 0.3 — it's a binary representation artifact:

6. Number Static Methods

Number.isInteger(value) — Integer Check

Number.parseFloat(str) / Number.parseInt(str, radix)

Number.prototype.toFixed(digits) — Format

7. The Math Object

Rounding

Min, Max, Range

Interview use: Range filters in data selection (#30).

Random

Other Useful Math Methods

Quick Reference

Interview Tips

• Use Number.isNaN, not isNaN — the global version coerces strings to numbers.
• Use Number.isFinite, not isFinite — same reason.
• Use Object.is for strict comparisons — handles NaN and signed zero.
• Clamp with Math.min(Math.max(v, min), max) — the one-liner for range constraints.

Related Articles

• Implement deepEqual()
• #30 — Implement Data Selection
• Data Types & Type Checking