Fdir – Faster Node.js glob alternative

栏目: IT技术 · 发布时间: 4年前

内容简介::zap:

Fdir – Faster Node.js glob alternative

The Fastest Directory Crawler for NodeJS

:zap: Extremely Fast: Nothing beats fdir in speed. It can easily crawl a directory containing 1 million files in < 1 second.

:bulb: Stupidly Easy: fdir only has 2 functions; sync and async for crawling the file system synchronously or asynchronously.

Zero Dependencies: fdir uses pure NodeJS fs & path namespaces. Nothing else is ever touched.

Astonishingly Small: < 1KB in size

:fire: All Node Versions Supported: fdir runs everywhere on all Node versions (within reason). And it is unsurprisingly fastest there too.

:bullettrain_side: Quickstart

You can install using npm :

$ npm i --save fdir

or Yarn:

$ yarn add fdir

It makes no difference to me.

const fdir = require("fdir");

// get all files in a directory synchronously
const files = fdir.sync("path/to/dir");

// or asynchronously
fdir.async("path/to/dir").then(/*blah blah blah*/);

And that's it.

:bar_chart: Benchmarks:

$ yarn benchmark

Specs:

  • Intel i7 7th Generation (7700HQ)
  • 16 GB of RAM
  • 256 GB SSD
  • OS: Manjaro Linux
  • Directory Size: 7386 files

Notes:

  • Some people asked that I benchmark no-op (without options) version of fdir . I did and found no performance difference. The results were identical. (I didn't include it here as it wasn't anything special.)

Node v13.11.0:

Synchronous (7386 files) Asynchronous (7386 files)
Fdir – Faster Node.js glob alternative Fdir – Faster Node.js glob alternative

Node v8.3.0:

Note: As latest version of rrdir doesn't support Node < 8, I had to use version 2.0.0. Everything else is fully updated.

Synchronous (7386 files) Asynchronous (7386 files)
Fdir – Faster Node.js glob alternative Fdir – Faster Node.js glob alternative

:fire_engine: API:

fdir is very small so there's not much to the API.

fdir.sync(string, Options): String[]

This is often the fastest way to get files. However, it will block the main "thread" so use it with caution with large directories.

fdir.async(string, Options): Promise<String[]>

Not always the fastest but works without blocking the street, so that's a plus.

Options

Ah, the options. Not many of them. At least not as many as I'd hoped for.

includeDirs: boolean

Whether to include directories in the array returned.

default: false

excludeBasePath: boolean

Whether to exclude the base path for each file.

default: false

searchFn: Function

Use this to filter out files.

Example:

fdir.sync("node_modules", {
  searchFn: path => path.includes(".git")
});

default: undefined

maxDepth: number

The max number of levels fdir should crawl before stopping. The lower the faster.

default: undefined (i.e. infinity)

isExcludedDir: Function

Use this to exclude particular directories from being crawled.

Example:

const isExcludedDir = path => path.includes(".bin");
fdir.sync("node_modules", { isExcludedDir });

default: undefined

ignoreErrors: boolean

Ignore errors while traversing the directory.

default: false

And that's it.

:interrobang: FAQs:

1. I looked at the code and there's nothing special. How is it so damn fast then?

Well, that's the whole point. fdir exists to prove to the "young" generation that you don't need to use special constructs or special methods to gain speed. Just a bit of patience and brains.

2. I found X library. I ran its benchmarks. It is faster than fdir !

Um. Well thank you for embarassing me (just joking). Do tell me the name of this library though. I will try to optimize fdir and reclaim the first spot :smile:

3. You are doing X and Y wrong! Do Z and it will improve performance!

Yes. And I should probably do A, B & C too. The point is, did you run benchmarks with these suggestions? If you did and saw significant improvements, thank you. Now go open a PR :laugh:

4. Why create this? What's the point?

I know you don't care. Fine. There's no point behind this. It's "just for fun". No, wait. Actually, I created this, first of all, for me. I needed fast directory access in another app of mine, so fdir came into being.

5. Why are all the other libraries so slow?

Because they did not spend enough time optimizing it. Most developers give readability and cool code more importance than actual performance and usability. I have seen a library claiming to be the fastest by inverting the benchmarks. Literally. Gave me quite the scare until I went and fixed the benchmark. It was actually one of the slowest. :O

6. How long did it take you to create this?

Ummm. Maybe 18 hours? Make it a day.

7. Are you looking for a job?

Am I? Well, are you offering a job? If yes, I am interested. :D

8. Why should I care?

You shouldn't. But here's my email in case you do: thecodrr[at]protonmail.com . Don't worry, I don't bite.

:information_source: Support

Would love if you buy me a cup of coffee right over here . Or just be, you know, polite and give me a star? Maybe even follow me?

LICENSE

Copyright (c) 2020 Abdullah Atta under MIT. Read full text here.


以上就是本文的全部内容,希望对大家的学习有所帮助,也希望大家多多支持 码农网

查看所有标签

猜你喜欢:

本站部分资源来源于网络,本站转载出于传递更多信息之目的,版权归原作者或者来源机构所有,如转载稿涉及版权问题,请联系我们

梦断代码

梦断代码

Scott Rosenberg / 韩磊 / 电子工业出版社 / 2008.06 / 49.00元

软件乃是人类自以为最有把握,实则最难掌控的技术。本书作者罗森伯格对OSAF主持的Chandler项目进行田野调查,跟踪经年,试图借由Chandler的开发过程揭示软件开发中的一些根本性大问题。. 本书是讲一事,也是讲百千事;是写一软件,也是写百千软件;是写一群人,也是写百千万人。任何一个在软件领域稍有经验的技术人员看完本书,必掩卷长叹:做软件难。...一起来看看 《梦断代码》 这本书的介绍吧!

随机密码生成器
随机密码生成器

多种字符组合密码

Markdown 在线编辑器
Markdown 在线编辑器

Markdown 在线编辑器

RGB HSV 转换
RGB HSV 转换

RGB HSV 互转工具