@lzwme/asmd-calc

简体中文

A short and concise JavaScript library for the four fundamental operations of arithmetic, which supports the addition, subtraction, multiplication and division calculation of floating-point precision.

Features

• Short and sharp. It only includes the calculation of addition, subtraction, multiplication and division. The core code is less than 100 lines. There is no other dependency.
• High execution performance. Only the four operations that can be accurately expressed by the front end are considered to realize simple logic and better execution performance than the open source library with rich and complex functions (see Benchmark for details).
• Accurate and stable. The accumulation of many years of practical experience in the development of several financial trading systems covers various common calculation use case error scenarios.

Not Applicable

• For numerical processing involving very large numbers that front-end can not be accurately represented.
• Complex mathematical scientific calculation.

These libraries are recommended:

• math.js
• decimal.js
• calculatorjs
• bignumber.js
• big.js/。

Install

# npm
npm i @lzwme/asmd-calc
# yarn
pnpm add @lzwme/asmd-calc
# pnpm
pnpm add @lzwme/asmd-calc

Useage

1. Calculation example using simple tool method

es module

import { add } from '@lzwme/asmd-calc';

console.log(add(0.1, 0.2, 0.3));
// => 0.6

commonjs

const calc = require('@lzwme/asmd-calc');

console.log(calc.add(0.1, 0.2, 0.3));
// => 0.6

2. Calculation example using chain operation

es module

import { AsmdCalc } from '@lzwme/asmd-calc';

const a = new AsmdCalc(0);
console.log(+a.add(0.1).add(0.2, 0.3));
// => 0.6

const b = new AsmdCalc(0.3);
b.add(0.1, 0.2).add(0.1).sub(0.1, 0.2).sub(0.1).div(0.3).mul(0.3);
console.log(+b);
// => 0.3
console.log(b.value);
// => 0.3

commonjs

const AsmdCalc = require('@lzwme/asmd-calc').AsmdCalc;

const calc = new AsmdCalc(0);
console.log(calc.add(0.1).add(0.2, 0.3));
// => 0.6

API

• add(...args); - Addition
• sub(...args); - Subtraction
• mul(...args); - Multiplication
• div(...args); - Division
• keepDotLength(value: number | string, precision: number, type: 'ceil' | 'round' | 'fround' | 'floor' | boolean = 'floor'): number | null; - Keep N decimal places
• toFixed(value: number | string, precision: number, type: 'ceil' | 'round' | 'fround' | 'floor' | boolean = 'floor'): string; - Similar to Number.prototype.toFixed, but fixed precision of the result
• getDecimalLen(num): number; - Get the decimal length
• toNonExponential(num): string; - Convert to string format of unscientific counting method

Development and testing

• Development

pnpm install
pnpm start

• Testing

npm test
npm run benchmark

• Build

npm run build

Benchmark

npm run benchmark

See：Benchmark

The following results are the time-consuming comparison of executing 10000 * N times on the same machine:

type/timeConst	jsRaw	asmd-calc	mathjs
add-10000	19.225ms	169.535ms	415.145ms
sub-10000	16.269ms	34.827ms	171.263ms
mul-10000	18.518ms	51.625ms	235.868ms
div-10000	27.025ms	79.504ms	300.706ms

Pre execution of 1000000 times and then stats time-consuming of execution of 10000 * N times:

type/timeConst	jsRaw	asmd-calc	mathjs
add-10000	7.768ms	155.836ms	362.819ms
sub-10000	8.339ms	25.147ms	155.611ms
mul-10000	9.995ms	35.685ms	224.357ms
div-10000	15.666ms	77.407ms	280.322ms

Benchmark Details

random-10000times:

type/timeConst	[jsRawCalc]	[asmdCalc]	[decimal]	[mathjs]
add	5.01ms	59.3607ms	95.3418ms	144.9646ms
sub	5.0866ms	9.3846ms	33.8741ms	62.0785ms
mul	5.4093ms	14.9623ms	54.6482ms	94.0914ms
div	7.1796ms	26.7458ms	72.871ms	117.2498ms

decimal-10000times:

type/timeConst	[jsRawCalc]	[asmdCalc]	[decimal]	[mathjs]
add	0.9363ms	5.597ms	12.3949ms	18.1254ms
sub	0.7557ms	7.9353ms	12.6472ms	18.1084ms
mul	0.4496ms	7.2964ms	14.6238ms	20.0657ms
div	0.4945ms	6.0127ms	15.5085ms	20.1288ms

integer-10000times:

type/timeConst	[jsRawCalc]	[asmdCalc]	[decimal]	[mathjs]
add	0.8038ms	0.5012ms	6.1372ms	13.3008ms
sub	0.4676ms	0.7482ms	6.4895ms	16.1568ms
mul	0.3157ms	0.5464ms	7.799ms	15.9254ms
div	0.614ms	0.4786ms	9.8204ms	17.2268ms

References

• 754-2008 – IEEE Standard for Floating-Point Arithmetic
• Floating Point Math – https://0.30000000000000004.com
• 确保前端 JavaScript 浮点数精度的四则运算方法

License

@lzwme/asmd-calc is released under the MIT license.

该插件由志文工作室开发和维护。