mdn-compat-table-scraper
We're converting our compatibility data into a machine-readable JSON format. ... Find out how you can help! -- MDN
About
A set of scraping tools to help convert MDN's HTML browser compatibility tables into JSON. So far, the best MDN compat-table scraper is Kayce Basques's MDN Crawler. This aims to improve and re-export these scraping tools for possible reuse and bundle them in a working scraper, called via bookmarklet.
Usage the bookmarklet to scrape old MDN compatibility tables
To use the scraper, go to https://skalt.github.io/mdn-compat-table-scraper/, save the bookmarklet to your bookmarks, then click it when you're on a MDN page that needs scraping.
Installation for reuse in other projects
npm install or yarn add [-S|-D] git+https://github.com/SKalt/mdn-compat-table-scraper.git
You can use any of the scraping functions by import {/* any non-trivial function */} from mdn-compat-table-scraper/src/index.js. Most functions require a JQuery api $ as their last arguement. The top-level scraper reqires window as well, to access window.location, which can be mocked.
Developing
git clone https://github.com/SKalt/mdn-compat-table-scraper.git
cd mdn-compat-table-scraper
# using npm
npm ci
npm run test
# using yarn
yarn install
yarn run test
Roadmap:
[x] gather the HTML of example pages (with an eye to diversity of table format)
[x] Write JSON validation functions (using mdn's json schemas)
[x] Bundle the scraper as a bookmarklet
[ ] improve docs