Skip to content

Latest commit

 

History

History
49 lines (40 loc) · 2.66 KB

README.md

File metadata and controls

49 lines (40 loc) · 2.66 KB

clean-code

Clean Code Principles

Intro

  • who do we write code for? for machines or for future people to understand?
  • there is only one valid metric for code review: no of swears per minute (show image)

Clean Code Principles

General ones:

  • SRP: Single Responsibility Principle: a module/class/function should have only one single reason to change
  • DRY: Don't Repeat Yourself. Extract duplicate code to reusable functions/classes.
  • KISS: Keep It Simple Stupid. Code should be very easy to read: almost in plain english. (choose proper names for functions, variables, constants, etc)
  • YAGNI: You Ain't Gonna Need It. Don't write code that you think you might need in the future. Write the minimum to achieve your task.
  • OCP: Open Closed Principle. A module/class should be Open to extensions and Close to modifications.
  • SOC: Separation Of Concerns. You should keep your code flow clear, structured and separated by concerns. Eg: Don't mix data validation with saving the data to the DB.

More spefific ones:

  • guardian statements + early returns: helps reduce the cyclomatic complexity and also the needed tests for a method.
  • methods length: keep it small, smaller.
  • passing parameters to methods: as few as possible (> 3 means SRP violation), no booleans (split instead in 2 methods)
  • utilities classes
  • DTOs: Data Transfer Objects
  • Comments: should explain why something is needed not how it works. The code should explain the hows.
  • Writing platform independent code. Eg: file path separators in Java instead of "/"
  • Constants: use them for magic numbers & hardcoded strings,
  • Avoid methods with side effects as much as possible: more like pure functions that operate on some input and return an output
  • Avoid swtich cases in multiple places in code. If switch case is needed, keep it in one place. Eg: Factory Pattern.

How to refactor

  • baby steps + often commits.
  • don't try to refactor all at once: you will fail, miserably. Instead, try to take a big mess and make it smaller, one step at a time.

When to refactor

  • on a daily basis: there is always room for small improvements
  • don't be afraid to change the code in order to make it more readable. Put emphasis on the need of having good unit tests.
  • be aware that too much refactoring can lead to too much abstractizations, which will make the code more harder to understand. Keep a fine balance between them.

What to refactor

  • take some code from the project and try to apply the principles together with the team

Conclusion

  • Continuous Refactoring: refactoring your code whenever you work on some feature/task.

Future discussion:

  • null vs exception,
  • concurrency & imutability
  • low coupling & high cohesion
  • architectural designs