Protected methods cannot be overridden
In Ruby, the main reason you would use protected instead of private is because you want different instances of the same class lineage to be able to access the method.
However, the rules around protected (callable only where self of the context is the same as the [...] method definition) means that protected method effectively cannot be overridden by subclasses. The redefinition of the method resets the protected access check to the subclass, so that instances of the parent class (or other subclasses) cannot call the method anymore.
Is the recommendation that using protected is just bad practice and should be avoided? Or is there a way to make the protected behavior aware of the parent method that is being overridden and keep the access check at the same level in the class hierarchy?
class Person def initialize(ssn) @ssn = ssn end def same?(other) tax_id == other.tax_id end protected def tax_id @ssn end end class HappyPerson < Person def shout 'yay!' end end # corporations are people now class Corporation < Person def initialize(ein) @ein = ein end protected def tax_id @ein end end bob = Person.new('000-00-0001') sally = HappyPerson.new('000-00-0002') acme = Corporation.new('00-0000001') puts bob.same? bob # true puts bob.same? sally # false puts sally.same? bob # false puts acme.same? bob # false puts bob.same? acme #=> protected method `tax_id' called ... (NoMethodError)
No data to display