hello-world-bot

A helloworld webhook nodejs sample for a SAP Conversation AI bot.

Introduciton

A helloworld webhook sample for a SAP Conversational AI bot with a webhook trigger as action. This simple sample how you reply with text message and a list template message. The nlp result by SAP Conversational AI is sent in the request body, in which you can know about the intent and entities etc as a result of your bot training. Usually, the intent indicates a service to be fulfiled by the bot.

For example, the end user is buying something through a conversation with your bot by saying "I want some pizza", once its intent and given contextual information has been understood by the bot, the bot will fulfils the service by invoking a RESTful call to SAP Business One or SAP Business ByDesign to place an order etc, and respond to the end user with the order number just placed.

To use this sample, please follow the steps below:

Prequisities

• You have followed this tutorial to build your first bot with SAP Conversational AI.

Download the project

git clone https://github.com/YatseaLi/hello-world-bot.git

Deploy the webhook project

Option 1: Local Deployment

Prequisites: You have NodeJS 6.12 or above run-time installed in your local machine.

1. Install the dependent node modules

   npm install

2. Start the app

   node hello_world_webhook.js
   

3. Publish the webhook url to public internet.
   If you don’t have a public server, or if you want to test your bot during development, ngrok is a very handy tool: It creates a public URL for you and forwards requests to your computer. Once you installed it, run

   ngrok http 5000
   

   And copy the Forwarding URL in HTTPS (https://XXX.ngrok.io) to your bot Settings (“Bot webhook base URL” field). All requests made to these URL will be forwarded to the port 5000 of your computer.

   As a result, you have obtain the webhook url for your bot.

Option 2: Deployment on Cloud Foundry, SAP Cloud Platform

Prequisitive: You have followed this tutorial to get started with Cloud Foundry

1. To avoid the naming conflict, edit the application name in manifest.yml. For example, "xxx-helloworld-bot"(xxx stands for your company namespace)

2. Open terminal and change directory to hello-world-bot

   cd hello-world-bot
   

3. Run the command to login to Cloud Foundry, SAP Cloud Platform,and deploy the app.

   cf login
   ...
   cf push
   

   Once the app is started, as a result, you now obtain the web hook url for your bot. for exmaple: https://xxx-helloworld-bot.<>/

Apply the webhook url to the skill of your bot

Open your bot on SAP Conversation AI

• Go to tab Build and create a skill for the target intent
• Setup the trigger as when the target intent is present.
• For action, use Call Webhook to reply, and fill in the webhook url as step of deployment.

Now test you bot, it will reply two messages:

• One as a text message "roger that"
• The other as a list of customer with sales amount and gross profit.
  For more reply templates, please refer to this online help

app.post('/', (req, res) => {
      console.log(req.body)
      //Todo: Please implement your service fulfilment for the intent here
      //And format the result of service fulfilment and send back the reply message.
      res.send({
    replies: [{
      type: 'text',
      content: 'Roger that',
    }, {
      "type": "list",
      "content": {
        "elements": [{
            "title": "C20000",
            "subtitle": "Sales Amount: 20000.00$\nGross Profit: 5000.00$",
            "buttons": [{
              "title": "View Details",
              "type": "BUTTON_TYPE",
              "value": "View Details"
            }]
          },
          {
            "title": "C30000",
            "subtitle": "Sales Amount: 30000.00$\nGross Profit: 8000.00$",
            "buttons": [{
              "title": "View Details",
              "type": "BUTTON_TYPE",
              "value": "View Details"
            }]
          }
        ],
        "buttons": [{
          "title": "View Chart",
          "type": "BUTTON_TYPE",
          "value": "View Chart"
        }]
      }
    }],
    conversation: {
      memory: {
        key: 'value'
      }
    }
  })

Here conclude the hello_world_webhook sample for a bot powered by SAP Conversational AI

Further todos for a complete webhook implementation of your own bot

1. Complete the service fulfilment in webhook endpoint for your own bot.
2. Format the result of service fulfilment above into reply back to user
   This sample only show one end point "/" for the webhook.
   Of course, you could have multiple endpoint, each serve one intent
   For instance, you have built an online shopping assistant bot with several inents, then you can endpoint as

• product-inquiry intent: app.post('/search-product', (req, res) => {}) ,which handle the product inquiry by searching the products from product catalog with given key words.
• add-to-cart intent: app.post('/add-to-cart', (req, res) => {}), which handle the add-to-cart intent by adding the given item to the shopping cart
• place-order intent: app.post('/place-order', (req, res) => {}) which handle the checkout the shopping cart and place the order to ERP.

License

The source code is under MIT license. Please kindly check the LICENSE. Here is to highlight that THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED. Therefore no support available.