Ruby Best Practices

Implementing Per-Object Behavior

An interesting aspect of Ruby is that not only can objects have per-class method definitions, but they can also have per-object behaviors. What this means is that each and every object carries around its own unique identity, and that the class definition is simply the blueprint for the beginning of an object’s lifecycle.

Let’s start with a simple irb session to clearly illustrate this concept:

>> a = [1,2,3]
=> [1, 2, 3]
>> def a.secret
>>   "Only this object knows the secrets of the world"
>> end
=> nil
>> a.secret
=> "Only this object knows the secrets of the world"
>> [1,2,3,4,5].secret
NoMethodError: undefined method 'secret' for [1, 2, 3, 4, 5]:Array
        from (irb):17
        from :0
>> [1,2,3].secret
NoMethodError: undefined method 'secret' for [1, 2, 3]:Array
        from (irb):18
        from :0

Here, using a familiar method definition syntax, we add a special method called secret to the array we’ve assigned to a. The remaining examples show that only a gets this new method definition. If the last one surprised you a bit, remember that most objects in Ruby are not immediate values, so two arrays set to [1,2,3] are not the same object, even if they contain the same data. More concisely:

>> [1,2,3].object_id
=> 122210
>> a.object_id
=> 159300

So when we talk about each object having its own behavior, we mean exactly that here. You may be wondering at this point what uses there might be for such a feature. An interesting abstract example might be to note that class methods are ...