Jupyter NodeJS: HTML Output (and React Demonstration)

config.js

module.exports = {
  extensions: {
    // clojurescript: require('./ext/clojurescript'),
    clojure: require('./build/ext/clojure'),
    clojurescript: require('./build/ext/clojurescript'),
    coffee: require('./build/ext/coffee'),
    babel: require('./build/ext/babel'),
    html: require('./build/ext/html')
  }
}

html.js

'use strict';

Object.defineProperty(exports, '__esModule', {
  value: true
});

exports['default'] = function (ctx, args, done) {
  done(null, {
    block: {
      html: function(ctx, args, code, out, done) {
        return ctx.rawEvaluate(code, out, function(e, output) { 
          if (e) return done(e);
          if (!output['text/plain']) return done(null, output);
          return done(null, {'text/html':eval(output['text/plain'])});   
	})
      } 
    } 
  });
};

module.exports = exports['default'];

usage.md

Instructions / usage

Add the html.js file to the package/build/ext/ directory, and update config.js as shown.

First run the following in your notebook:

%load_ext html

Then, at the start of any block in which you will output html, place this:

%%html
// your code here, object to display on final line as usual

And here's a full example, rendering a simple react component:

(don't forget to npm install react && npm install react-dom from notebook folder first)

Thanks. I got this working, but there's an error in the instructions.
%%load_ext html should be %load_ext html in usage.md

Thanks, updated to fix that.

And I'll repost the link to your other helpful note here, where you indicated that another somewhat simpler method for using HTML already exists, for anyone interested:

notablemind/jupyter-nodejs#21 (comment)