@steve02081504/bigfloat

A simple JavaScript library for arbitrary-precision arithmetic, also known as "bigfloat". This implementation is inspired by elc.

It allows you to work with very large numbers and high-precision decimals without floating-point inaccuracies. It represents numbers as fractions, enabling precise calculations.

Features

• Infinite Precision: Numbers are stored as fractions (numerator and denominator) for maximum precision.
• Basic Arithmetic: Supports addition, subtraction, multiplication, division, modulus, and power operations.
• Comparisons: Accurately compare numbers.
• String Conversion: Handles conversion to and from strings, including support for repeating decimals (e.g., 0.3[3]).
• Expression Evaluation: Includes a built-in eval function to parse and compute mathematical expressions from a string.

Installation

npm install @steve02081504/bigfloat

Usage

Basic Operations

import { bigfloat } from '@steve02081504/bigfloat';

const a = bigfloat('10').div(bigfloat('3')); // 10/3
const b = new bigfloat('0.5'); // 1/2

console.log(a.toString()); // '3.[3]'
console.log(b.toString()); // '0.5'

// Addition
const sum = a.add(b);
console.log(sum.toString()); // '3.8[3]' (10/3 + 1/2 = 23/6)

// Multiplication
const product = a.mul(b);
console.log(product.toString()); // '1.[6]' (10/3 * 1/2 = 10/6)

// Negative numbers
const c = new bigfloat('-123.456');
console.log(c.neg().toString()); // '123.456'

Repeating Decimals

The library can parse and represent repeating (recurring) decimals using bracket notation [].

import { bigfloat } from '@steve02081504/bigfloat';

// Parse a repeating decimal
const repeating = bigfloat('0.[1]'); // 1/9
console.log(repeating.toString()); // '0.[1]'

const result = repeating.mul(bigfloat('9'));
console.log(result.toString()); // '1'

Expression Evaluation

Use the static bigfloat.eval() method to evaluate mathematical expressions from a string.

import { bigfloat } from './main.mjs';

const result1 = bigfloat.eval('1.2 * (3.4 + 5.6)');
console.log(result1.toString()); // '10.8'

const result2 = bigfloat.eval('(1/3) + (2/7)');
console.log(result2.toString()); // '0.[428571]' (13/21)

const result3 = bigfloat.eval('2**100');
console.log(result3.toString()); // '1267650600228229401496703205376'

Supported operators in eval:

• +, -, *, /, %, ** (power)
• ==, !=, <, >, <=, >=
• &&, ||, ! (logical)
• Parentheses () for grouping.

API

new bigfloat(value)

Creates a new bigfloat instance. value can be a string, number, or another bigfloat instance.

Arithmetic

• .add(other)
• .sub(other)
• .mul(other)
• .div(other)
• .mod(other)
• .pow(other)

Comparison

• .equals(other)
• .lessThan(other)
• .greaterThan(other)
• .compare(other): Returns 1 if greater, -1 if less, 0 if equal.

Other Methods

• .neg(): Returns the negated value.
• .abs(): Returns the absolute value.
• .floor(): Returns the floor of the number.
• .toString(): Converts the number to its string representation, using [] for repeating decimals.

Static Methods

• bigfloat.eval(expression): Evaluates a mathematical expression string.
• bigfloat.fromString(string): Creates a bigfloat from a string.