Ruby has this handy and convenient way to share instance variables by using keys like
attr_accessor :var
attr_reader :var
attr_writer :var
Why would I choose attr_reader or attr_writer if I could simply use attr_accessor? Is there something like performance (which I doubt)? I guess there is a reason, otherwise they wouldn't have made such keys.
You may use the different accessors to communicate your intent to someone reading your code, and make it easier to write classes which will work correctly no matter how their public API is called.
class Person
attr_accessor :age
...
end
Here, I can see that I may both read and write the age.
class Person
attr_reader :age
...
end
Here, I can see that I may only read the age. Imagine that it is set by the constructor of this class and after that remains constant. If there were a mutator (writer) for age and the class were written assuming that age, once set, does not change, then a bug could result from code calling that mutator.
But what is happening behind the scenes?
If you write:
attr_writer :age
That gets translated into:
def age=(value)
@age = value
end
If you write:
attr_reader :age
That gets translated into:
def age
@age
end
If you write:
attr_accessor :age
That gets translated into:
def age=(value)
@age = value
end
def age
@age
end
Knowing that, here's another way to think about it: If you did not have the attr_... helpers, and had to write the accessors yourself, would you write any more accessors than your class needed? For example, if age only needed to be read, would you also write a method allowing it to be written?