I’m creating an inheritor from ComponentBoundsConstrainer, and trying to override checkBounds(), but hitting a problem. Normally I can copy and paste the body of the function I’m overriding into my class, and then alter it, but I can’t do that in this case.
For one, all the minima and maxima for height and width are private. Not a big deal, I can use the accessor functions for those as replacements.
checkBounds() also uses minOffTop and related private member variables, and I can’t access these in my public inheritor, as there are no accessor functions.
Is there a way around this? In my own classes I tend to make member variables protected rather than private, so that inheritors can access member variables directly, without having to use accessors. Is that bad design?
Can’t you just call ComponentBoundsConstrainer::checkBounds() at the start of your overloaded method? Are you copying and pasting because you didn’t know about being able to do that in c++, or because of some other reason?
In the end I did do that. I’m still curious about the more general question, however: are there advantages to having member variables be private rather than protected in classes that you expect will be overridden.
I can see pros and cons to both philosophies, so getting your perspective might help me improve my code down the road.
I don’t really have a philosophy on it that I could explain, it’s more just a feeling for which bits of a class I don’t want people to mess with!
The main reason it’s best to keep things private is that if someone’s subclass expects a variable to behave in a certain way, and then you change the way the base class is implemented, it could silently break all the derived code.
