#9. Keep functions short and cohesive

-
A function should have one specific purpose. And it should achieve its purpose by means of a cohesive structure.

A function can be structured in one of these ways:
  • A sequence of actions.
  • A condition.
  • A repetition of certain actions.
These control flow structures can be combined to form more sophisticated units. For example, a function may be a sequence of actions, with some of them being executed only under some conditions, and some other being loops. A certain degree of composition may be useful, but if you're not very strict, you'll soon find yourself writing 50-line functions which lack any structure.

Software scientists have long attempted to measure the complexity of code. Cyclomatic complexity measures the number of linearly independent paths through a program's source code. The higher cyclomatic complexity a function has, the harder it is to test.

If your function is a condition, you may need to test it in two different scenarios (paths): when the condition is true, and when it is false. If it contains four conditions, suddenly the scenarios become 2^4, that is, 16. If it is a combination of nested conditions with some loops spread out here and there, it will be a testing and maintenance nightmare. Some of these long and overly functions contain bugs ever since they begin to exist. Nobody has found them yet, because they are hidden in a path which requires five different conditions to be met. But they will appear, eventually.

Real life software problems are usually complex, but you should divide that compexity among different, simple software elements, each of them doing its job in an effective and easy to test way. The function is the smallest software element: you can't write unit tests for something smaller than a function. Complex "parts" of a potentially long function deserve to be in their own functions, with clearly defined semantics and which can be tested independently.

Bibliography

[McConnell 2004]
Steve McConnell: Code Complete, 2nd Edition, Microsoft Press, 2004.

This book has an interesting section about complexity in software, 34.1: "Conquer Complexity" (pages 837-839), inside its valuable chapter 34: "Themes in Software Craftmanship".

Comments

Post a Comment

Popular posts from this blog

Welcome to The Deep Blue C++

-
Image
During my years as a software engineer, I have learnt quite a lot of things - and I keep learning every day. I learn from practice, discussions with workmates, reading... I especially learn from mistakes: code that doesn't do what I expect, code that doesn't do anything at all, code that does it but with some occasional crash or two... If you're a programmer, you know what I'm talking about. What I've learnt has substantially changed the way I write code. In my first years as a programmer I tried to avoid mistakes, but I hadn't made enough of them yet to discover the best ways to prevent them. Now I know better. I know there are some things that bring you into trouble very quickly. Other things will silently wait for the worst moment - maybe months or years later, to come to the surface in the form of even more original problems. Disorganized code, uninitialized variables, pointer arithmetics, implicit casts, syntactic overloading, inheritance in which ...
Read more

#2. Make all type conversions explicit

-
Assignments, function calls and expressions in general should not contain implicit type conversions. All type conversions should be explicit. Type conversions should be isolated in their own line, which should consist only of the assignment of the result of the type conversion to a variable of the target type. A statement should not mix type conversion with other operations. For example, if you need to convert an object to one of another type to pass it as a function parameter, do not do it in the function call itself; instead, do it in a separate instruction in the previous line. Type conversion occurs when an entity of type A is, well, converted to an entity of type B. This may occur in function calls, expressions using operators, or assignments. An explicit type conversion is one which you can read in the code. An implicit type conversion is one which does not appear in the source code, but is done automatically when the software is executed. Making all type conversi...
Read more

#13. Organize your code in classes

-
The minimal unit of code is a statement . The minimal unit of code which can be tested and reused is a function . Functions are rarely practical all by themselves; they are usually organized in modules which offer a set of operations which make as a whole. The class is the main semantical unit in object oriented programming. A class is an abstract data type with an associated implementation. As Bertrand Meyer puts it in Object Oriented Software Construction, 2nd Edition : Classes should be the only modules. These concepts will not be obvious in the integrated development environment you use to create your code. Usually, these IDEs let you create projects, and inside them, they let you add units or files. You know, of course, the C++ syntax; you know there are two keywords, class and struct , that let you define a type with its associated data members (fields) and operations (functions). You are probably aware of the difference between class and struct : the default visibili...
Read more