Note: I do not ask whether or not this is reasonable thing to do or if this is good design. I'm just asking if this is well-defined behaviour and if the results are as expected.
I came upon a following class hierarchy:
struct A
{
virtual void foo() = 0;
};
struct B: public A
{
void foo() override
{
std::cout << "B::foo()\n";
}
};
struct C: public B
{
virtual void foo() = 0;
};
struct D: public C
{
void foo() override
{
std::cout << "D::foo()\n";
}
};
int main()
{
A* d = new D;
d->foo(); //outputs "D::foo()"
// A* c = new C; // doesn't compile as expected
}
Is this code well defined? Are we allowed to override definition with pure-specifier?
[class.abstract/5] of the current draft Standard:
[Note: An abstract class can be derived from a class that is not abstract, and a pure virtual function may override a virtual function which is not pure. — end note]
The very same note is included even in the C++11 Standard. So, the answer is yes, it is valid.