C++ - Why set object to null after deleting? [duplicate]

Question

I am looking at the following piece of linked list code I found online:

void DeleteAfter(Node **head){
      if(*head==NULL){
            return;
      }else{
            Node *temp = NULL;
            temp = (*head)->next;
            (*head)->next = (*head)->next->next;
            delete temp;
            temp=NULL;
      }
}

I am not that skilled with C++, so this could be a bad question, but why is temp being set to NULL after being deleted? Is this a necessary step?

Answer 1

It's unnecessary. Some people make a habit of doing this even when it has no result. An aggressive compiler optimizer will eliminate this code, so it doesn't actually do any harm. But I would have written:

void DeleteAfter(Node *head) {
  if (head) {
    Node *next = head->next;
    if (next) {
      head->next = next->next;
      delete next;
    }
  }
}

Note I eliminated a useless level of indirection and added a check to make sure there's a "node after" to delete.

The rationale for the habit is that if a pointer always refers to either a valid object or is null, you can rely on null checks as equivalent to validity checks.

For this reason, Ada, a language often used in safety critical systems, initializes pointers to null and defines its delete equivalent operator to set its argument null automatically. Your C++ is simulating this behavior.

In practice, the value of this discipline is not what you'd hope. Once in a long while it prevents a silly error. One nice thing, however, is that debugger displays of pointer content make sense.