Parse HTML metadata (including OpenGraph and TwitterCard) in NodeJS and Browser compatible JavaScript. It's about 12.7 kB gzipped and support a large HTML encoded entities.
$ npm i --save opengraph-metadata-parserArguments: (html, baseUrl)
Return value: Object
Description:
Take
htmlsource-code of a page along itsbaseUrlwhich is the website main URL with the proto (as https://) but without the trailing slash. It return an Object with key that are independant and may beundefined,String, and sometimeInteger,ArrayofString.
Possible returned keys are:
titletype(from "og:type")url(page URL in a convenient format)descriptiondeterminent(from "og:determinet")locale(as "EN", "FR", ...)sitename(from "og:site_name")keywords(Array of String)icon(URL, the highest resolution one)image(URL)image_type(mime types)image_width(Integer)image_height(Integer)image_altvideo(URL) /video_type/video_width/video_height/video_altaudio(URL) /audio_title/audio_type
See src/index.js for more informations
import parsePage from 'opengraph-metadata-parser';
const {locale} = parsePage("<html lang="FR_fr"></html>", "https://mainsite.fr");
console.log(locale);
// print: FR- It may detect tag in most cases without any issue. It also decode HTML5 chars.
- It may parse tag that must not be parsed as commented one. It shouldn't be problematic as no serious website use comments in production code.
- As it DO NOT use any (virtual) DOM, it may be a lot faster that other similar project.
- For furthere information, read source code.
I did some basic test to ensure a minimum quality. They are not so much enforced but may be enough for the most cases. See test folder for further informations.