Code Smell 21 - Anonymous Functions Abusers

Functions, lambdas, closures. So high order, nondeclarative, and hot.

Problems

• Maintainability

• Testability

• Code Reuse

• Implementation Hiding

• Debugging

Solutions

1. Wrap functions/closures

2. Reify algorithms in method object / Strategy

Sample Code

Wrong

Right

Detection

• Closures and anonymous functions are very useful to model code blocks, promises etc. So It'd difficult to tear them apart.

Tags

• Primitive

• Abuser

Conclusion

Humans read code. Software works ok with anonymous functions but maintainability is compromised when multiple closures are invoked.

Relations

Code Smell 06 - Too Clever Programmer

Maxi Contieri ・ Oct 25 '20 ・ 1 min read

#codenewbie #tutorial #beginners

Credits

Photo by Roman Mager on Unsplash

---

Object-oriented programming increases the value of these metrics by managing this complexity. The most effective tool available for dealing with complexity is abstraction. Many types of abstraction can be used, but encapsulation is the main form of abstraction by which complexity is managed in object-oriented programming.

Rebecca Wirfs-Brock

Comments

[Magne]

Given that "Humans read code", as you said, I find the first example wayyy easier to read than the second. The second example has over 2x as many lines and over 3x as many characters... I don't know if introducing classes was the right solution here, as it adds indirection and complexity. To me, it thus seems that Right is Wrong and Wrong is Right.

To modify the first example so you solve your 5 stated problems, you could simply:

1. Rename sortFunction() → bubbleSort().

2. Inside it (the sort function), rename fn → compare, for clarity, so the if sentence reveals the intent: if (compare(arr[j], arr[j+1]))

3. Name the anonymous function before passing it into the (newly renamed) sort function:
   

const compare = (a,b) => { return a > b };
sorted = bubbleSort(scores, compare);

I do agree that anonymous functions can and often are abused, though. But mostly because they are often injected everywhere (typically obfuscating param lists), without the developer having taken the the time to name and declare it (which would simplify the params list, and also reveal the function's intent better).

[DarkWiiPlayer]

Problems

• Maintainability
• Testability
• Code Reuse
• Implementation Hiding
• Debugging

1. That's very subjective. To one developer might find it easier to maintain an object-oriented codebase, while others might prefer working on functional code.
2. Normally, an anonymous function is tested indirectly along with the surrounding code.
3. Re-using an anonymous function is trivial in most cases: just extract it into a named function when you find yourself needing it a second or third time.
4. There's two interfaces in your first example:
   • the higher order sortFunction
   • presumably the interface of the outer function surrounding the snippets For both of them, the interface is completely separated from its implementation, just as it'd be with an object. If you extract the comparator function in a later refactoring step, it will also share this property.
5. There's no real difference between a debugger telling you "error in function X" and "error in anonymous function on line Y in file Z".