Quality-Dashboard

This project is born out of the need to centralise reports that are generated in a typical CI/CD flow: unit tests, coverage, integration tests, UI tests, ...

This is made to be as simple and as light as possible.

Installation

The recommended aproach is to deploy the services as containers. A deployment is composed of the following services:

• Server: this is the backend service holding the data and the reports
• UI: this is the web frontend to display the reports

Depending on your preferred deployment target, You will find examples of deployment files in the examples directory:

• Docker Compose: [examples/quality-dashboard-docker-compose]
• Kubernetes: [examples/quality-dashboard-kubernetes]

Deployment Options

Some aspects can be customized. Parameters can be change with environment variable.

• Server:
  • Data volume: the server image stores the data in the /opt/data/
  • The BASEPATH: the root path of the APIs. default: /api
  • The AUTH_TOKEN_VALIDITY: validity of the authentication token in seconds. default: 3600
  • The API_CORS: the alloed origin for the web interface (CORS). default: *
• UI:
  • The BASEPATH: the root path of the web interface. default: /
  • The BASEPATH_SERVER: The URL of the server. default: /api

APIs

Get Reports

GET <API_URL>/reports/

Add Report

POST <API_URL>/reports/<group_name>/<project_name>/<version_name>/<report_name>/<processor_name>

The following are arbitrary value: group_name, project_name, version_name, report_name The following is a value that should match one of the exising processsor: processor_name This function should upload a file:

• The field is: report
• The file is either a .html file or a .tar.gz file

Examples:

# Send Archive
curl -X POST \
    -F [email protected] \
    http://localhost<API_URL>/reports/qa-dash/server/1.0.0/unit-test-coverage/lcov-coverage

# Send File
curl -X POST \
    -F [email protected] \
    http://localhost<API_URL>/reports/qa-dash/server/1.0.0/unit-test/jest-html-reporter

# JSON
curl -X POST \
    -d '{"link":"https://github.com/DidierHoarau/quality-dashboard", "success": 10, "error": 9, "warning": 8, "total": 27, "coverage": 80 }' \
    -H "Content-Type: application/json" \
    http://localhost<API_URL>/reports/qa-dash/server/1.0.0/test-json/json

# JSON (alternate)
curl -X POST \
    http://localhost<API_URL>/reports/qa-dash/server/1.0.0/test-json-alt/json?data_json=%7B\"success\"%3A10,\"error\"%3A9,\"warning\"%3A8,\"total\"%3A27,\"coverage\"%3A80%7D

# JSON and Html Report
curl -X POST \
    -F report=@"./report.html" \
    http://localhost<API_URL>/reports/qa-dash/server/1.0.0/test-json-file/json?data_json=%7B\"success\"%3A10,\"error\"%3A9,\"warning\"%3A8,\"total\"%3A27,\"coverage\"%3A80%7D

Processors

The analysis of reports is done by Processors. Processors are JavaScript files that must be put in the following folders:

• /opt/data/processors: For custom user-defined processors
• /opt/app/plugins/processors: For build-in report

The processor to be used is defined in the Add Report API.

A processor must have the following characteristic:

• Its name is the file name (without extension)
• It should export a function called analyse
• The analyse function will be passed a string for the location of the report (uncompressed)
• This function should return a Promise
• This Promise should resolve an object with the following fields (all fields are optional):
  • link: Relative link to the html page
  • success: Number of successful items
  • warning: Number of not fully successful items
  • error: Number of failed items
  • total: Total number of items
  • coverage: String representing the coverage

Example:

module.exports = {
  analyse: async reportFolder => {
    console.log(`Analysing ${reportFolder}`);
    return {
      link: 'report.html',
      success: 0,
      warning: 0,
      error: 0,
      total: 0,
      coverage: '0%'
    };
  }
};

There are built-in processors for:

• Jest (HTML output)
• JSON (given as POST data)
• LCov (HTML output)
• Popeye (HTML output)
• Trivy (HTML output)