dtsse-ardoq-adapter

Provides an integration to Ardoq in order to maintain the tech stack. The relationship for Ardoq components is like so:

erDiagram
    "VCS Platform" ||..|{ "Code Repositories" : hosts
    "HMCTS Application" ||..o{ "Code Repositories" : maintains
    "Software Dependency" }o..|{ "Code Repositories" : "depends upon version"

Loading

This application is a NodeJS application that runs in CFT and hosts an API that is called by HMCTS applications and the CNP Pipeline to update Ardoq with its dependencies.

flowchart LR
    A[GitHub Workflow] & B[CNP Jenkins Pipeline] -- Dependencies --> C[dtsse-ardoq-adapter]
    C -- Dependencies --> D[Ardoq]

Loading

Getting Started with Integration

There are currently 3 methods of integration. You will need to know your Ardoq Application ID for all of them.

The Ardoq Application ID can be found by logging into the Ardoq web portal, finding your application ini the 'HMCTS Applications' workspace and using the component id for the application.

1. Using the CNP Jenkins Pipeline

If your application is deployed using the standard CNP Pipeline then you need to onboard your application by raising a Pull Request such as this one: https://github.com/hmcts/cnp-jenkins-config/pull/977/files

Once this config PR is merged, your teams application data shout start appearing in Ardoq after your next master build.

2. Using the API directly

You will need to request an API key from the DTSSE team who can be found on the MoJ slack workspace in #rse-dev-tools.

You will need to POST the following payload to the /api/dependencies.

{
  "vcsHost": "Github HMCTS",
  "hmctsApplication": "YOUR_APPLICATION_ID_HERE",
  "codeRepository": "${{ github.event.repository.name }}",
  "encodedDependencyList": "",
  "parser": "yarn",
  "language": "",
  "languageVersion": ""
}

The full OpenAPI spec is included in this repository.

Some fields require additional clarification:

• language - The main language used in the application eg. java, nodejs, python etc.
• languageVersion - The version of the language used in the application eg. 11, 14, 3.8 etc.
• encodedDependencyList - The base64 encoded list of dependencies for the application. See the Supported Build Tools section for more information on how to get this data.

3. Using GitHub Workflows

Below is a sample GitHub workflow. for a yarn application.

name: Maintain Ardoq Tech Stack
run-name: ${{ github.actor }} Maintaining Ardoq Tech Stack. 🚀
on:
  push:
    branches:
      - master
      - main
jobs:
  Maintain-Ardoq-Tech-Stack:
    runs-on: ubuntu-latest
    steps:
      - run: echo "🔎 The name of your branch is ${{ github.ref }} and your repository is ${{ github.repository }}."
      - name: Check out repository code
        uses: actions/checkout@v3

      - run: echo "💡 The ${{ github.repository }} repository has been cloned to the runner."
      - name: List dependencies
        run: |
          cat yarn.lock | base64 > yarn.lock.b64
        id: deps

      - name: Prepare request payload
        run: |
          echo -n '{
            "vcsHost": "Github HMCTS",
            "hmctsApplication": "48f7e1661af3e8c64059f99f",
            "codeRepository": "${{ github.event.repository.name }}",
            "encodedDependencyList": "' > payload.json
          cat yarn.lock.b64 | tr -d '\n' >> payload.json
          echo -n '",
            "encodedDependencyListOther": "' >> payload.json
          cat package.json.b64 | tr -d '\n' >> payload.json
          echo -n '",
            "parser": "yarn",
            "language": "' >> payload.json
          grep -E '^FROM' Dockerfile | awk '{print $2}' | awk -F ':' '{printf("%s", $1)}' | tr '/' '\n' | tail -1 >> payload.json
          echo -n '",
            "languageVersion": "' >> payload.json
          grep -E '^FROM' Dockerfile | awk '{print $2}' | awk -F ':' '{printf("%s", $2)}' >> payload.json
          echo -n '"
            }' >> payload.json

      - name: compress body
        run: |
          gzip payload.json

      - name: debug compression
        run: |
          ls -lah payload.json.gz

      - name: post deps to ardoq app
        run: |
          curl -w "%{http_code}" --location --request POST '${{ secrets.ARDOQ_ADAPTER_URL }}?async=true' \
            --header 'Authorization: Bearer ${{ secrets.ARDOQ_ADAPTER_KEY }}' \
            --header 'Content-Type: application/json' \
            --header 'content-encoding: gzip' \
            --data-binary '@payload.json.gz'

      - run: echo "🍏 This job's status is ${{ job.status }}."

Supported Build Tools

The following sections of the supported build tools will describe how to get the correct payload to submit to the different endpoints for your built tool.

Gradle

./gradlew -q dependencies > $PWD/deps.log will provide the full output of the dependencies used but needs to be base64 encoded before sending to the API. The following command will do this for you: cat $PWD/deps.log | base64 > $PWD/deps.log.base64 This data then needs posting to the /api/gradle/<your repository name> endpoint.

Maven

For maven projects you can run mvn dependency:tree > $PWD/deps.log

This data then needs posting to the /api/maven/<your repository name> endpoint.

Dotnet

dotnet list package --format json will output the json content you need to then post to /api/dotnet/<your repository name>.

NPM

To integrate your project when using npm, you simply need to post the contexts of your package-lock.json file to /api/npm/<your repository name>. This works for both version 1 and 2 of the npm lock file.

Yarn

To integrate your project when using yarn, you simply need to post the contexts of your yarn.lock file to /api/yarn/<your repository name>.

PIP

To integrate your project when using pip, you simply need to post the contexts of your requirements.txt file to /api/pip/<your repository name>.

Maintaining this application

Prerequisites

Running the application requires the following tools to be installed in your environment:

• Node.js v12.0.0 or later
• yarn
• Docker

Running the application

Install dependencies by executing the following command:

$ yarn install

Bundle:

$ yarn webpack

Run:

$ yarn start

The applications's home page will be available at https://localhost:8506

Running with Docker

Create docker image:

  docker-compose build

Run the application by executing the following command:

  docker-compose up

This will start the frontend container exposing the application's port (set to 8506 in this template app).

In order to test if the application is up, you can visit https://localhost:8506 in your browser. You should get a very basic home page (no styles, etc.).

Developing

Code style

We use ESLint alongside sass-lint

Running the linting with auto fix:

$ yarn lint --fix

Running the tests

This template app uses Jest as the test engine. You can run unit tests by executing the following command:

$ yarn test

Here's how to run functional tests (the template contains just one sample test):

$ yarn test:routes

Security

CSRF prevention

Cross-Site Request Forgery prevention has already been set up in this template, at the application level. However, you need to make sure that CSRF token is present in every HTML form that requires it. For that purpose you can use the csrfProtection macro, included in this template app. Your njk file would look like this:

{% from "macros/csrf.njk" import csrfProtection %}
...
<form ...>
  ...
    {{ csrfProtection(csrfToken) }}
  ...
</form>
...

Healthcheck

The application exposes a health endpoint (https://localhost:8506/health), created with the use of Nodejs Healthcheck library. This endpoint is defined in health.ts file. Make sure you adjust it correctly in your application. In particular, remember to replace the sample check with checks specific to your frontend app, e.g. the ones verifying the state of each service it depends on.

License

This project is licensed under the MIT License - see the LICENSE file for details