Concepts

​

How It Works

Whenever a team member opens a pull request, Kody performs a Code Review directly in your code management tool, such as GitHub or GitLab. Kody analyzes the code, provides detailed feedback, and highlights potential issues, just as a human reviewer would. This ensures code quality and speeds up the review process.

Additionally, you can use the command @kody start-review directly in Git to have Kody initiate a code review on any pull request. This command bypasses branch and naming configurations, so Kody will review any PR where the command is used.

​

Code Review Settings

Customize Kody’s Code Review to align with your team’s specific needs.

​

Configuration Hierarchy

Kody allows you to configure settings at two levels: Global and Repository-specific. Here’s how the hierarchy works:

​

Global Settings

Global settings apply to all repositories by default. If no repository-specific configuration is set, the global configuration will be used for that repository. This is a great way to establish a baseline for all your projects.

​

Repository-specific Settings

If you configure settings for a specific repository, those settings will override the global configuration for that repository. This allows for tailored configurations based on the needs of individual repositories.

​

How It Works

• No repository-specific configuration: The global settings will apply.
• Repository-specific configuration exists: The settings for that repository will take precedence over the global settings.

​

Example Use Case

• You enable Automated Code Review globally and configure global analysis types like Security and Refactoring.
• For a specific repository (e.g., kodus-service-slack), you decide to disable Automated Code Review or enable additional analysis types like Error Handling.
• In this case, the configuration for kodus-service-slack will apply only to that repository, while all other repositories will follow the global settings.

​

Automatic Review

Choose whether you want reviews to trigger automatically or only when you issue the command @kody start-review in a pull request.

​

Types of Analysis

Select the types of analysis you’d like Kody to perform. Options include:

• Performance Optimization: Identifies improvements for code efficiency, performance, and resource usage.

  // Original code
  let result = [];
  for (let i = 0; i < items.length; i++) {
    result.push(items[i] * 2);
  }
  
  // Optimized code using map (more efficient and readable)
  const result = items.map((item) => item * 2);
  

• Security: Detects vulnerabilities and improves code security.

  // Original code
  function getUserData(username) {
    return db.query(`SELECT * FROM users WHERE username = '${username}'`);
  }
  
  // Improved code with enhanced security (using prepared statements)
  function getUserData(username) {
    return db.query("SELECT * FROM users WHERE username = ?", [username]);
  }
  

• Error Handling: Enhances how errors and exceptions are managed.

  // Original code
  const data = JSON.parse(input);
  
  // Improved code with error handling
  try {
    const data = JSON.parse(input);
  } catch (error) {
    console.error("Error parsing JSON:", error);
  }
  

• Refactoring: Restructures code to improve readability, maintainability, or modularity.

  // Original code
  function calculateDiscount(price, discount) {
    if (discount) {
      return price - (price * discount) / 100;
    } else {
      return price;
    }
  }
  
  // Refactored code to reduce duplicated logic
  function calculateDiscount(price, discount = 0) {
    return price - (price * discount) / 100;
  }
  

• Maintainability: Enhances code to simplify future maintenance and extension.

  // Original code
  function processOrder(order) {
    // order proce