How to Integrate IBM Watson with CircleCI

Set Up IBM Watson API Credentials

• Sign in to your IBM Cloud account. If you don’t have one, you can create a free account.

 

• Go to the IBM Watson catalog and choose the Watson service you wish to use (e.g. Watson Assistant, Watson Natural Language Understanding).

 

• Click on "Create" to set up your chosen service.

 

• Navigate to your service instance and find the "Manage" tab to get your API credentials, which include the API Key and Service URL.

Install Required Developer Tools

• Ensure that you have git installed on your machine. You can download it from the official site.

 

• Install CircleCI CLI to interact with the CircleCI platform. You can install it using Homebrew:

  brew install circleci
  

 

• Set up your local development environment for the language in which your application is written (e.g., Node.js, Python).

Create a CircleCI Configuration File

• In your project directory, create a folder named `.circleci` and inside it, create a file called `config.yml`.

 

• Write the configuration necessary to build and test your project. Here's a basic example for a Node.js project:

  version: 2.1
  
  jobs:
    build:
      docker:
        - image: circleci/node:latest
      steps:
        - checkout
        - run: npm install
        - run: npm test
  
  workflows:
    version: 2
    build_and_test:
      jobs:
        - build
  

 

• Modify the configuration file to include an additional step for integrating with IBM Watson, such as calling an endpoint using cURL or any client library available for your service.

Store IBM Watson API Credentials in CircleCI

• Open your CircleCI project at CircleCI dashboard or create a new project if you haven't linked one yet.

 

• Go to "Project Settings" and navigate to "Environment Variables".

 

• Create new environment variables for `IBM_API_KEY` and `IBM_SERVICE_URL`, using the values obtained from IBM Cloud. This ensures that sensitive information is not hard-coded into your project files.

Configure CircleCI to Interface with IBM Watson

• Modify your test or build script within `config.yml` to incorporate IBM Watson API calls. Here is an example of a script using cURL:

  - run:
      name: Call IBM Watson API
      command: |
        curl -X POST -u "apikey:${IBM_API_KEY}" \
        --header "Content-Type: application/json" \
        --data "{\"text\": \"Hello IBM Watson\"}" \
        "${IBM_SERVICE_URL}/v1/analyze?version=2020-08-01"
  

 

• Ensure the proper handling of API responses in your application logic to proceed with subsequent steps in the CI/CD pipeline.

Test and Validate Your Integration

• Commit and push your changes to the version control repository linked to CircleCI. This will trigger a new build based on the configuration you specified.

 

• Check the CircleCI dashboard to verify that the integration was successful and that the API calls are being executed as expected.

 

• Analyze any errors or failures in the CircleCI logs to troubleshoot and refine your implementation.

Implement Advanced Configurations as Needed

• Explore more advanced configurations like parallelism for running jobs faster, or larger machine executors if your Watson tasks are computation-intensive.

 

• Consider adding more complex automations such as notifying team members on build failures which involve Watson API responses. Integrations with tools like Slack can be beneficial here.