How do you determine the need to eliminate code duplication?

Code duplication is an anti-pattern that reduces maintainability, readability, and flexibility of a system. I identify the need to eliminate duplication when I notice:

1. Repeated blocks of code: Identical or very similar logic fragments that appear in multiple places.
2. Same functionality: Different parts of the system perform the same task but in different ways.
3. Problems when making changes: Changing logic in one place requires similar changes everywhere with duplication, increasing the risk of errors and effort.
4. Understanding complexity: It is difficult for the reader to understand the true meaning because the logic is scattered and repeated.
5. Violation of the DRY principle (Don't Repeat Yourself): The code violates the basic development principle that each piece of knowledge should have a single, unambiguous, and authoritative representation in the system.

Tools and approaches for identifying duplication:

• Manual code analysis: Visual review of the codebase.
• Static code analyzers: Using tools like PyLint, flake8, bandit, or specialized tools like Radon or Duplicate Code Detector, which can automatically find duplicates.
• Refactoring: During refactoring, opportunities to eliminate duplication are often discovered.
• Team discussion: Colleagues can point out repeated code blocks.

After identifying duplication, I decide to eliminate it by extracting the repeated logic into functions, classes, or modules.