Thread: Help Me Understand Destructor Function in Vector

You should show the definition of the Object class. The code is probably wrong as it is unlikely that the destructor for myString should be explicitly involved in the destructor of the Object class.

If the Object class is actually a mix-up with the Node class, i.e., by Object you mean Node, then this is definitely wrong: when a Node object is destroyed, the myString member will already be destroyed. If there is any work for the Node class destructor to do, then it likely has to be manually destroying the objects that the pointers in the vector point to, but whether this is true or not depends on whether you want the Node to own those other objects.

Okay, here's your code with proper formatting:

Code:

classObject
{
public:
    Object();
    ~Object();
    //other function prototypes here
private:
    vector<Node*> myString;
    Object*Last;
    Object*First;
};

Object::~Object()
{
    myString.~vector<Node*>(); 
    //Someone Help me understand this
}

So, perhaps you have a main function like this:

Code:

int main()
{
    Object x;
}

x is default constructed, so it has a vector<Node*> member variable named myString that is empty. At the end of the main function, x is destroyed, invoking its destructor. We see this line:

Code:

myString.~vector<Node*>();

The above is an explicit invocation of the destructor for the member variable named myString, i.e., on that line, the member variable named myString is destroyed. It is good for this member variable to be destroyed as after all we are destroying the Object object that owns it, so we want to destroy its members too. The trouble is that this member variable will already be destroyed at the end of the Object destructor, hence this explicit invocation of the destructor for the member variable named myString is a mistake that results in an attempt to destroy the member variable twice, resulting in undefined behaviour.