In Slack2Watson(1) and (2) I created a chatbot and integrated it into Slack Slash Commands. Now, let’s integrate the same chatbot into Facebook Messenger. To accomplish this I need to create a Facebook Application, add the Messenger Platform to my Facebook Application, create a Webhook, have admin access to a Facebook Page to generate a Page Access Token so I can send and receive messages send to the Facebook Page, and create the Node-RED flows to integrate the Facebook Messenger with the IBM Watson Conversation.
- Create a Flow to Verify the Request for Webhook Edits in Node-RED,
- Create an Endpoint for the Redirect URL of the Webhook in Node-RED,
- Create a Facebook Application for the Messenger Platform,
- Enable Webhooks Integration with Node-RED,
Create a Flow to Verify the Request for Webhook Edits in Node-RED
To prepare the setup and configuration of the Facebook Application, the Messenger platform and Webhooks to enable a chatbot in Facebook Messenger, I will first create the Node-RED flows to implement the required server workflow.
The first flow is to verify the endpoint for the setup of the Facebook Application, using the ‘hub.challenge’ token.
- Go to your Node-RED application on Bluemix at http://<username>-nodered-slackapp.mybluemix.net/,
- Click the ‘Go to your Node-RED flow editor’ button,
- If you’re not logged in yet, log in now,
- Add a new flow tab and rename the flow ‘Facebook Messenger’,
- To verify your endpoint during setup of your Webhook, or when you update an existing topic subscription of your Webhook, Facebook sends a GET request. The request will include:
hub.challenge â€” a random string
- Slash Commands in Slack,
- Watson Conversation service to create a ChatBot,
- Node-RED to configure the Application Flow,
Using a chatbot, I want to automate the following scenario:
- user: Hello
- chatbot: Hello. Where are you?
- user: hi, i am at the Rubin Museum.
- chatbot: I love the Rubin Museum. Shall I give you some recomendations of my personal favorites?
- user: oh yes, I would love some recommendations.
- chatbot: do you like sculpture, paintings or ornaments?
- user: i prefer paintings!
- chatbot: ok, here are my favorite paintings at the Rubin Museum: a, b, c
- Create the Watson Conversation for the Rubin Museum Scenario,
- Test the Conversation for the Rubin Museum Scenario,
- Setup Slash Commands in Slack,
- Create the Node-RED Flow to Watson Conversation,
- Add Token Validation of the Slack Request to Node-RED Flow,
Create the Watson Conversation for the Rubin Museum Scenario
- Go to IBM Bluemix and login to your account,
- To create the Watson Conversation, click the ‘Create Service’ button, which will take you to the Catalog, or
- Go to the Catalog, Under ‘Services’ filter by ‘Watson’, or in the catalog browse to the Watson section,
- Click the ‘Conversation’ service,
- Agree or change the ‘Service name’ and the ‘Credentials name’, and click the ‘Create’ button,
- Under the ‘Credentials’ tab, you will find the username, password, and workspace ID that you need later to configure access to the conversation service,
- Click the green ‘Launch tool’ button,
- First create a workspace, click ‘Create’, name the workspace ‘Watson2Slack-Workspace’, and click ‘Create’,
- You are now in your Conversation workspace, and you should see 3 tabs: Intents, Entities, and Dialog,
An extended version of this tutorial, adding a Cloudant NoSQL Database and D3js data visualization, is available here.
- Bluemix account
- Github account
- Create the StarterApp
- Add the Toolchain or Continuous Integration (CI)
- REST API Primer
- Authentication in REST API
- Getting the Bluemix Configuration
- Add the AlchemyData News API Client
- Add a Web Form
- Implement the AlchemyData News API Request
- Create Authorization Header for Basic Auth
1. Create the StarterApp
You must have Python installed. Check to see if you have Python installed from the commandline:
Table of Contents:
- Create a Starterapp
- Git Clone and Setup
- Create Additional Folders and Files
- Commit and Push Changes to Repository, Build and Deploy
- Create an AlchemyAPI service
- Create a Cloudant NoSQL service
- Add the Basic Workflow for Request-Response
- Get News using AlchemyAPI
- Create Helper Functions
- Save Responses in Cloudant
- Parse Response for D3js
- Draw the Line Chart in D3js
You can import AlchemyAPI requests into Postman with this Postman collection.
The source code for the application can be viewed or cloned from Github.
1. Create a Starterapp
- Go to Catalog > Boilerplates
- Click the ‘Python Flask’ starterapp
- For name enter <username>-newssentiment
- Go to Overview
- Under ‘Continuous Integration’ click ‘Add GIT Repo and Pipeline’ to add a DevOps platform, select ‘Populate the repo with the starterapp packageand enable Build & Deploy pipeline’ > Click Continue > Click ‘CLOSE’.
- Click ‘EDIT CODE’.
- The very first time you login to the ‘DevOps’ environment you will need to pick a username for the ‘DevOps’ environment.
- In the left menu of icons, click the top folder icon, and click ‘Git URL’ to copy the Git repository url.
- If you prefer to edit in the online editor in Bluemix, click ‘EDIT CODE’ button and then click the second pencil icon in the left menu of icons.
- I will continue to work on localhost instead.
Note: this API needs to be updated to v2. Identify is now a resource on the Language Translation API.
Making an HTTP REST API POST request in Java is in principle as simple this:
Request request = Request.Post(serviceURI)
Executor executor = Executor.newInstance();
String response = executor.execute(request).returnContent().asString();
This code, uses the Apache Fluent API, which exposes only the essentials of the Apache HTTPClient.
The Authorization header is required for authentication by the Watson API, and is a base64 encoded string.
String auth = username + ":" + password;
String basicAuthorization = "Basic "+ Base64.encodeBase64String(auth.getBytes());
The POST request is “x-www-form-urlencoded” and in Java consists of a List<NameValuePair>:
List<NameValuePair> params = new ArrayList<NameValuePair>();
You can easily add very cool IBM Watson APIs to your application. You need to sign up for Bluemix, create an application and bind APIs or services to the app. This will create credentials (username and password) for each API or service.
Create a Bluemix acccount
First if you have not signed up for a Bluemix account yet, go to http://ibm.biz/bluemixnyc and Sign Up.
Create an App
Once you have confirmed your account, go to the Catalog and choose any Runtime. We will only use the runtime to bind services to, which will give us API credentials. For example, name the app ‘my-app1’.
Add a Service
Now, to add Services either go to Catalog > Services, and add a service to your app, or go to your app’s admin page. Bluemix is based on Cloud Foundry, so apps in Bluemix are listed under ‘CF Apps.’ Go to Dashboard > CF Apps > ‘my-app1’, and click the ‘Add a Service or API.’ I will add Personality Insights to my app in this example. Edit the required fields and click the Create button. Bluemix might ask you to restage your app. Now we’re ready to call the Personality Insights API.
Call the Watson API
To make a REST call to a Watson API, you must add an Authorization header with a Base64 encoded “Basic :” string. To generate the Basic Authorization header, use the colon as a separator for username and password and Base64 encode the Authorization string.
String username = "username from your service credentials";
String password = "password from your service credentials";
String s = username + ":" + password;
byte b = s.getBytes();
String enc = Base64.encodeBase64String(b);
String authorization = "Basic "+enc;
To call the API with the above credentials use the following curl statement from command line.
curl -X POST -H "Authorization: Basic BzAvEATz4UYqKfU5JA66GSIwKNgdATGiADR8YiWaOSH2Purd6tedhKUJaPa3KLshLq==" -H "Content-Type: text/plain" -H "Accept: application/json" -H "Accept-Language: en" -H "headers: true" -d 'put your text here' https://gateway.watsonplatform.net/personality-insights/api/v2/profile
For the API reference details go here
This curl command will return a response object that you can visualize with the Visualize Personality Insights API.