Object-oriented theory you may not care or know about, but are doing

October 04, 2016

The law of demeter

A method m in a class A should not send a message to another object unless that object is one of the following:

An instance variable used in the method m;
A parameter to the method m;
An object created in the method m;
A global variable (public static variable) used in the method m.

One of the commonest examples is not exposing List<T> as properties, but IEnumerable<T> and providing methods to Add and Remove from the underlying list.

SOLID principles

I’ve asked about these so many times in interviews and I still can’t remember all the words behind the acronym - the point isn’t to memorise some words but rather understand the ideas behind them. The most important parts (for me) are having a single responsibility for an object, IoC/DI and having many interfaces rather than just one.

Single responsibility. “Do one thing and do it well”
Open/closed principle. Less important unless you are doing a lot with inheritence. Modification over extension of classes.
Liskov substitution principle. I’ve seen this abused with 8 class inheritence chains for a class ultimately is about performing a search. When abused, you end up unsure what the result of your override will do because of the mess above it.
Interface Segregation Principle. This ties into single responsibility - an example of it going wrong is where you might implement an interface and be forced to put empty implementations for half the methods.
Depdency inversion principle. If you’re using a DI framework like Structuremap then you’ll be following this.

Forks and Cascades

I’ve deliberately avoid LINQ to remove any confusion:

Fork

var microsoft = new Company();
Person john = microsoft.GetPerson("john");
int age = john.Age;

Cascade

  var microsoft = new Company();
  Person john = microsoft.GetPerson("john");
  int age = microsoft.GetAge("john");

Inheritence vs Composition

Fairly self-explanatory: inheritence is inheriting from a base class while composition is allowing properties of your classes to be filled, e.g. MyCompany.Person. Most of the time you will use composition, and DI you get to use it via Setter injection, or constructor injection which then fill properties.

object-orientedcsharp

I'm Chris Small, a software engineer working in London. This is my tech blog. Find out more about me via Github, Stackoverflow, Resume