A toy example might be needed to make it clearer what you are after, but I guess you want a class A which has as two of its properties a variable PropA and an object B.
If that's right, then it wouldn't be good OOP to have a call to B modify PropA, even if that is possible.
It would be better to make PropA a function (of A) that simply calls the function of B (call it "GiveMePropA") that returns the value you want PropA to have. If GiveMePropA is expensive to compute, you might just have A call it whenever needed and store its value.
0 Comments
Sign in to comment.