notdilbarsl.github.io

Website Link : https://notdilbarsl.github.io/

Assignment 2: Project Enhancements

Tools Used

• ESLint: For linting the codebase.
• SonarJS: To identify code duplication and code smells.
• Jest: For system testing.

How to Use the Tools

• Linting: Run npx eslint . in the terminal.
• Testing: Execute npm test in the terminal.

Changes Made

1. Initial Linting and Code Refactoring:

   • During the initial linting process, I discovered numerous issues related to unused variables and imports. This was because functions were spread across various files without proper import/export statements.
   • To resolve this, I transitioned to JavaScript modules and ensured that all import/export statements were correctly implemented. This change resolved approximately 100 of the 400 errors flagged by the linter.
   • Addressed code duplication and complexity issues using SonarJS. Reconfigured ESLint to flag errors for cognitive and cyclomatic complexity exceeding a threshold of 10.

2. Addressing Code Duplication and Complexity::

   • I used the SonarJS plugin to identify and eliminate code duplication.
   • I reconfigured the ESLint settings to flag errors if cognitive or cyclomatic complexity exceeded a threshold of 10. This led to several improvements and refactoring of code:
     • Cyclomatic Complexity:
       • Received an error in movePlayer.js due to high cyclomatic complexity. I refactored the code by breaking it into smaller functions such as calculateDirection.js, checkLoss.js, and checkWin.js, which simplified the logic and reduced the complexity.
     • Cognitive Complexity:
       • Encountered issues in generateRandomPath.js, which was created to simplify randomMinePlacement.js. I rewrote this file and created a new file, generateDirection.js, which contains a single function to manage the complexity more effectively.

3. Unit Testing:

   • I initially wrote tests using Jasmine and nyc (Istanbul) for coverage reports but faced compatibility issues with ES6 dependencies. Consequently, I switched to Jest for testing.
   • Achieved the following coverage metrics with Jest:
     • Statements: 97.14%
     • Branches: 91.56%
     • Lines: 97.37%
     • Functions: 94.87%
   • These metrics significantly exceeded the required 50% coverage.

4. CI/CD Pipeline:

   • Implemented a GitHub Actions pipeline to automate the workflow:
     • Linting: Runs on every push to the main branch.
     • Testing: Executes unit tests if linting passes.
     • Deployment: Deploys to GitHub Pages if tests pass and coverage exceeds 50%.

These improvements were crucial in enhancing the quality and maintainability of the project.