Skip to content

Ruby#

Source code

This is a minimal server application built for Ruby with Sinatra that allows:

It internally uses LiveKit Ruby SDK.

Running this application#

Download the tutorial code:

git clone https://github.com/OpenVidu/openvidu-livekit-tutorials.git

To run this server application, you need Ruby installed on your device.

  1. Navigate into the server directory 
    cd openvidu-livekit-tutorials/application-server/ruby
  2. Install dependencies 
    bundle install
  3. Run the application 
    ruby app.rb

Info

You can run any Application Client to test against this server right away.

Understanding the code#

The application is a simple Ruby app using the popular Sinatra web library. It has a single file app.rb that exports two endpoints:

  • /token : generate a token for a given Room name and Participant name.
  • /livekit/webhook : receive LiveKit webhook events.

Let's see the code of the app.rb file:

app.rb
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
require 'sinatra'
require 'sinatra/cors'
require 'sinatra/json'
require 'livekit' # (1)!
require './env.rb'

SERVER_PORT = ENV['SERVER_PORT'] || 6080 # (2)!
LIVEKIT_API_KEY = ENV['LIVEKIT_API_KEY'] || 'devkey' # (3)!
LIVEKIT_API_SECRET = ENV['LIVEKIT_API_SECRET'] || 'secret' # (4)!

set :port, SERVER_PORT # (5)!

register Sinatra::Cors # (6)!
set :allow_origin, '*' # (7)!
set :allow_methods, 'POST,OPTIONS'
set :allow_headers, 'content-type'
set :bind, '0.0.0.0' # (8)!
  1. Import livekit library
  2. The port where the application will be listening
  3. The API key of LiveKit Server
  4. The API secret of LiveKit Server
  5. Configure the port
  6. Enable CORS support
  7. Set allowed origin (any), methods and headers
  8. Listen in any available network interface of the host

The app.rb file imports the required dependencies and loads the necessary environment variables (defined in env.rb file):

  • SERVER_PORT: the port where the application will be listening.
  • LIVEKIT_API_KEY: the API key of LiveKit Server.
  • LIVEKIT_API_SECRET: the API secret of LiveKit Server.

Finally the application configures the port, sets the CORS configuration for Sinatra and binds the application to all available network interfaces (0.0.0.0).

Create token endpoint#

The endpoint /token accepts POST requests with a payload of type application/json, containing the following fields:

  • roomName: the name of the Room where the user wants to connect.
  • participantName: the name of the participant that wants to connect to the Room.
app.rb
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
post '/token' do
  body = JSON.parse(request.body.read)
  room_name = body['roomName']
  participant_name = body['participantName']

  if room_name.nil? || participant_name.nil?
    status 400
    return json({errorMessage: 'roomName and participantName are required'})
  end

  token = LiveKit::AccessToken.new(api_key: LIVEKIT_API_KEY, api_secret: LIVEKIT_API_SECRET) # (1)!
  token.identity = participant_name # (2)!
  token.add_grant(roomJoin: true, room: room_name) # (3)!

  return json({token: token.to_jwt}) # (4)!
end
  1. A new AccessToken is created providing the LIVEKIT_API_KEY and LIVEKIT_API_SECRET.
  2. We set participant's identity in the AccessToken.
  3. We set the video grants in the AccessToken. roomJoin allows the user to join a room and room determines the specific room. Check out all Video Grants.
  4. Finally, we convert the AccessToken to a JWT token and send it back to the client.

The endpoint first obtains the roomName and participantName parameters from the request body. If they are not available, it returns a 400 error.

If required fields are available, a new JWT token is created. For that we use the LiveKit Ruby SDK:

  1. A new AccessToken is created providing the LIVEKIT_API_KEY and LIVEKIT_API_SECRET.
  2. We set participant's identity in the AccessToken.
  3. We set the video grants in the AccessToken. roomJoin allows the user to join a room and room determines the specific room. Check out all Video Grants.
  4. Finally, we convert the AccessToken to a JWT token and send it back to the client.

Receive webhook#

The endpoint /livekit/webhook accepts POST requests with a payload of type application/webhook+json. This is the endpoint where LiveKit Server will send webhook events.

app.rb
36
37
38
39
40
41
42
43
44
45
46
47
post '/livekit/webhook' do
  auth_header = request.env['HTTP_AUTHORIZATION'] # (1)!
  token_verifier = LiveKit::TokenVerifier.new(api_key: LIVEKIT_API_KEY, api_secret: LIVEKIT_API_SECRET) # (2)!
  begin
    token_verifier.verify(auth_header) # (3)!
    body = JSON.parse(request.body.read) # (4)!
    puts "LiveKit Webhook: #{body}" # (5)!
    return
  rescue => e
    puts "Authorization header is not valid: #{e}"
  end
end
  1. Get the Authorization header from the HTTP request.
  2. Create a new TokenVerifier instance providing the LIVEKIT_API_KEY and LIVEKIT_API_SECRET. This will validate the webhook event to confirm it is actually coming from our LiveKit Server.
  3. Verify the Authorization header with the TokenVerifier.
  4. Now that we are sure the event is valid, we can parse the request JSON body to get the actual webhook event.
  5. Consume the event as you whish.

We need to verify that the event is coming from our LiveKit Server. For that we need the Authorization header from the HTTP request and a TokenVerifier instance built with the LIVEKIT_API_KEY and LIVEKIT_API_SECRET.

If the verification is successful, we can parse the request JSON body and consume the event (in this case, we just log it).

Remember to return a 200 OK response at the end to let LiveKit Server know that the webhook was received correctly.

From production to a local server#

When developing locally pointing to a production deployment and webhooks events are required by your application, you might face issues because OpenVidu cannot access your local server.

To receive webhooks from OpenVidu on your local machine, you need to expose your local server to the internet. This exposure allows OpenVidu to send webhooks directly to your local server.

The following images illustrate the difference between an unreachable local server and a reachable local server:

Unreachable local server

Unreachable local server

Reachable local server

Reachable local server

Exposing your local server to the internet is a common practice when developing applications locally. Tools like Ngrok, LocalTunnel, LocalXpose and Zrok can help you achieve this.

These tools provide you with a public URL that forwards requests to your local server. You can use this URL to receive webhooks from OpenVidu. For information on how to add this URL as the webhook URL in the OpenVidu deployment, refer to the following documentation:

  • Configure webhooks for an OpenVidu Local deployment. Learn more.
  • Configure webhooks for an OpenVidu Single Node deployment. Learn more.

  • Configure webhooks for an OpenVidu Elastic On-Premises deployment. Learn more.

  • Configure webhooks for an OpenVidu Elastic AWS deployment. Learn more.

  • Configure webhooks for an OpenVidu High Availability On-Premises deployment. Learn more.

  • Configure webhooks for an OpenVidu High Availability AWS deployment. Learn more.