Skip to content

Latest commit

 

History

History
68 lines (50 loc) · 2.1 KB

README.md

File metadata and controls

68 lines (50 loc) · 2.1 KB
Raw

magnet-html-parser Build Status

Extracts key document metadata from an HTML string.

const parser = require('magnet-html-parser');
parser.parse(html, url).then(result => ...);

Results

The parser will attempt to return the following metadata:

  • title - A title of the page
  • description - A description of the page
  • image - An image that best represents this page
  • icon - The site's icon
  • siteName - The name of the site which this page belongs
  • themeColor - A theme color of the site

What metadata is returned completely depends on how the page is marked-up. We cannot guarantee any results.

Hash Fragments

If the given URL has a hash fragment extension, magnet-html-parser will attempt to fetch an element with the given ID and return a title, description and image from that scope.

  • title - The textContent of the first heading (<h1>,<h2>,<h3>, ...) descendant of the fragment element
  • description - The textContent of the first <p> descendant of the fragment element
  • image - The src of the first <img> descendant of the fragment element.
  • scoped - Will be true when content was found within the given fragment scope.

Adaptors

An adaptor is a custom parser that performs some specific metadata extraction for a matched URL pattern. The library comes with some built in adaptors to cope with some common sites, but you are able to add more.

const parser = require('magnet-html-parser');
const adaptor = parser.adaptor;

parser.use(adaptor.extend({
  pattern: /example.com/,

  title() {
    var mainHeading = $('h1').first();
    return mainHeading.text();
  },

  description() {
    var firstParagraph = $('p').first();
    return firstParagraph.text();
  },

  image($, data) {
    return $('.main-image').attr('src');
  }

  ...
}));

As you can 'extending' the base adaptor, you only need to define the properties you wish to customise, all other properties will be parsed using the default logic.