https://github.com/masylum/cheer-reader

A port of readability.js for cheerio
https://github.com/masylum/cheer-reader

Last synced: 6 months ago
JSON representation

A port of readability.js for cheerio

• Host: GitHub
• URL: https://github.com/masylum/cheer-reader
• Owner: masylum
• License: other
• Created: 2024-06-28T10:34:34.000Z (about 2 years ago)
• Default Branch: main
• Last Pushed: 2024-10-23T08:09:04.000Z (over 1 year ago)
• Last Synced: 2025-01-22T12:32:40.843Z (over 1 year ago)
• Language: HTML
• Size: 5.13 MB
• Stars: 13
• Watchers: 1
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • Changelog: CHANGELOG.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE.md
  • Code of conduct: CODE_OF_CONDUCT.md

Awesome Lists containing this project

README

          
![cheer-reader](https://github.com/masylum/cheer-reader/blob/main/img/cheer-reader.png?raw=true)

This library is a port of the [readability.js](https://github.com/mozilla/readability) library

that uses [cheerio](https://cheerio.js.org) instead of native DOM APIs.

The goals of this port are the following:

-   **Portability**: Most runtimes and Web Workers don't support DOM apis.

    By using cheerio, we ensure that you can run readability wherever you want.

    If your goal is to run it in the main thread of the browser, you can stick to the original

    implementation.

-   **Performance**: For non-native runtimes, we rely in JSDOM to simulate native DOM APIs. JSDOM

    does a lot besides parsing HTML, and is pretty heavy and slow. **cheerio** instead is lightweight

    and fast. Overall, I've experienced that this port performs 6-8x times faster and consumes a lot

    less memory than the original one. You can even use [htmlparser2](https://github.com/fb55/htmlparser2)

    if you need more performance.

-   **Compatibility**: In order to avoid using JSOM, you could use readability's [JSDOMParser](https://github.com/mozilla/readability/blob/main/JSDOMParser.js),

    [Happy Dom](https://github.com/capricorn86/happy-dom) or [Linkedom](https://github.com/WebReflection/linkedom).

    I had compatibility problems using those alternative implementations, but **cheerio** worked flawlesly even

    with the most broken html documents that you can find on the internet. You even have the option to use

    [htmlparser2](https://github.com/fb55/htmlparser2) instead of **cheerio**'s default [parse5](https://github.com/inikulin/parse5)

    for less strict parsing.

-   **Extensability**: The original implementation is pretty hard to read and maintain. While I didn't deviate

    much from it, I tried to modernize the code base a little bit. I hope people can help me out, specially with

    the `_grabArticle` gigantic method with nested while loops. I also saw a lot of issues on the repo for top pages

    that didn't seem taken care of. It is my intention to try to fix those as I move forward.

The implementation aims to be the same and it keeps the same test set to ensure backwards compatibility.

## Installation

`cheer-reader` is available on jsr:

```bash

npx jsr add @paoramen/cheer-reader

```

## Basic usage

To parse a document, you must create a new `Readability` object from a DOM document object, and then call the [`parse()`](#parse) method. Here's an example:

```javascript

import { Readability } from 'cheer-reader'

const $ = load('html>
yo