Instance variables in ruby may be a bit confusing when first learning Ruby, especially if you are accustomed to another OO language like Java.
You cannot simply declare an instance variable.
One of the most important things to know about instance variables in ruby, apart from the notation with an @ sign prefix, is that they spring into life the first time they are assigned to.
class Hello
def create_some_state
@hello = "hello"
end
end
h = Hello.new
p h.instance_variables
h.create_some_state
p h.instance_variables
# Output
[]
["@hello"]
You can use the method Object#instance_variables to list all instance variables of an object.
You normally “declare” and initialize all the instance variables in the initialize method. Another way to clearly document which instance variables that should be publicly available is to use the Module methods attr_accessor (read/write), attr_writer (write) and attr_reader (read). These methods will synthesize different accessor methods for the listed instance variable.
class Hello
attr_accessor :hello
end
h = Hello.new
p h.instance_variables
h.hello = "hello"
p h.instance_variables
# Output
[]
["@hello"]
The instance variable still isn’t created until it’s assigned to using the synthesized Hello#hello= method.
Another important issue, like kch described, is that you need to be aware of the different contexts active when declaring a class. When declaring a class the default receiver (self) in the outermost scope will be the object that represents the class itself. Hence your code will first create a class instance variable when assigning to @hello on the class level.
Inside methods self will be the object on which the method is invoked, hence you are trying to print the value of an instance variable with the name @hello in the object, which doesn’t exists (note that it’s perfectly legal to read a non existing instance variable).