Question about the "cast" function

[JasonM]

I don’t think I understand what we are doing by using the cast function. At 2:40 in the lecture, Sam says that we are using it to “check that the USceneComponent is actually a UPrimitiveComponent”.

Before posting this question I did a little bit of searching and what I found was all the explanations described casting as a “conversion” of one type to another, which seems to be different from what says stated.

Regardless whether it’s a check or a conversion, I’m not understanding why we can state that a "USceneComponent " is actually a “UPrimitiveComponent” because they aren’t, right? Does this have to do with the Inheritance/Composition concept, were you are saying an object “is a”, or “has a”?

Thanks!

[DanM]

You should search dynamic casting to get better search results about what is happening here.

You’re checking whether the object at the address is a UPrimitiveComponent. As a USceneComponent can point to objects of derived types as well (so yes re: inheritance).

A UPrimitiveComponent is a USceneComponent so it can be safely upcasted. however the reverse isn’t true. So down casting is not implicit as it’s not safe to do so. i.e.

struct Animal {};
struct Dog : Animal{};

Dog dog;
Animal* animal = &dog; // 👍
Dog* dog_ptr = animal; // ❌ (even though we can see it is directly above)

That is what Cast does, it performs runtime type checking. It will check the type of the object at the address; if it is of that type you specify you get the same address back, otherwise nullptr is returned.

Small demo: Compiler Explorer

Note: dynamic_cast is how you do it in standard C++ and it requires the class to be polymorphic which means having at least one virtual function.