beatport-scraper
v3.0.22
Published
Scraper for Beatport data
Downloads
16
Readme
Beatport Scraper
A scraper for music data, using Beatport as source
Installation
The package is available on npm
, yarn
and gpr
(github package registry).
NPM & Yarn
$ npm i beatport-scraper # npm
$ yarn add beatport-scraper # yarn
Github Package Registry
Create a .npmrc
with the line below:
@wesselsmit:registry=https://npm.pkg.github.com
Install the package in the CLI
$ npm i @wesselsmit/beatport-scraper
Usage
This scraper gets data from an artist or label Beatport account.
url
string: URL of Beatport account to scrape. (required)log
boolean: Log progress in console. (optional, defaults to false)raw
boolean: Return data unformatted. (optional, defaults to false)
Example
The example scrapes all data from Seven Lions's Beatport account.
const scrape = require('@wesselsmit/beatport-scraper') // using gpr
const scrape = require('beatport-scraper') // using npm or yarn
const config = {
url: 'https://www.beatport.com/artist/seven-lions/241780',
log: true, // optional
raw: false // optional
}
scrape(config)
.then(data => console.log(data))
Disclaimer
The nature of web scraping is that when the HTML/website changes, the web scraper will inevitably fail. Beatport has every right to change/improve their UI as they see fit. This scraper relies on the DOM structure and selectors to navigate the website and identify data. Changes to Beatport's website can cause the scraper to break!
Maintaince
Status: working (last checked: August 15th, 2020)
As explained in disclaimer, this scraper might break in the future if Beatport has changed their website.
To make troubleshooting and maintaince easier the scraper relies on as little as possible DOM selectors and DOM specific code. All code specific to the Beatport website is either in the scraper.js script or in the selectors.js module.
Recommended first steps for troubleshooting/maintaince:
- check if the DOM selectors in the selectors.js module are still up to date with the website.
- check if the lines marked with
//! Subject to change
comments in scraper.js are still up to date with the website.