express-js.md

ExpressJS

For voice apps in prototyping stage, we recommend using a local webserver and a service like jovo webhook or bst proxy. This way, you can easily update your app without having to upload it to a server or AWS Lambda every time.

• Webhook Configuration
• Deploy to a Server
  • Verification
  • Run Server

Webhook Configuration

Jovo uses the express framework for running a server. Here is how the part of index.js, which is used to run the app on a webhook, looks like:

'use strict';

const { Webhook, ExpressJS } = require('jovo-framework');
const {app} = require('./app.js');

if (process.argv.indexOf('--webhook') > -1) {
    const port = process.env.PORT || 3000;

    Webhook.listen(port, () => {
        console.info(`Local server listening on port ${port}.`);
    });

    Webhook.post('/webhook', async (req, res) => {
        await app.handle(new ExpressJS(req, res));
    });
}

You can either run your server locally, or deploy to a webhosting service.

Deploy to a Server

The jovo webhook URL is only meant for prototyping purposes and can't be used in production. Follow the steps here to find out how to deploy your voice app to a production server.

Verification

When you want to deploy your code to a webserver other than AWS Lambda, you need to verify that Alexa Skill requests are actually coming from Amazon.

For this, Jovo uses a package called alexa-verifier-middleware, which can be accessed by switching one line of the configuration in index.js:

// Use this
const { WebhookVerified, ExpressJS } = require('jovo-framework');

// Instead of this
const { Webhook, ExpressJS } = require('jovo-framework');

To make use of it, please install it like so:

$ npm install alexa-verifier-middleware

Run Server

To run the server, use the following command:

$ node index.js --webhook