Here is a basic program for insertion at the beginning of a LL.
I have a fair idea of double pointers. Please will you tell me what exactly is the need of using a double pointer to head here in this code? node * head would have sufficed I believe.Code:void insert( node **head, int item) { node *ptr; ptr=(node *) malloc(sizeof(node)); ptr->info= item; if( *head == NULL) ptr->next = NULL; else ->ptr->next = *head; *head = ptr; }
Consider this line:
suppose head was a node* instead of a node**. How would you rewrite the above line to reflect this, and then what would happen (or not happen)?Code:*head = ptr;
Look up a C++ Reference and learn How To Ask Questions The Smart WayOriginally Posted by Bjarne Stroustrup (2000-10-14)
What does a *node_name hold? Since a node is a structure of multiple data types, what does *node_name fetch? I know int *c, *c would fetch the integer stored at address pointed to by c.
So you have a pointer named ptr, right. Now imagine you did this:
node **head = &ptr;
Now *head == ptr. Does this help? It should make sense to you that a pointer is also stored in a location with a memory address.
Why exactly do we need double pointers as arguments here? Can't the same be achieved without it? Eg. instead of what you suggested,
node * head = &ptr;
head== ptr;
works the same way, or don't they? Please explain.
Why don't you run it to found out yourself what is happening?
Code - functions and small libraries I use
It’s 2014 and I still use printf() for debugging.
"Programs must be written for people to read, and only incidentally for machines to execute. " —Harold Abelson
A pointer is declared by naming the thing it points to followed by an asterisk (*). If the pointer points to a pointer then the type of that is a "double pointer".Why exactly do we need double pointers as arguments here?
If you tried to compile:
node *head = &ptr;
you should get at least a warning. The location being assigned and the object being assigned are not the same type.
'node * head' would not suffice. You are modifying 'head' within your insert() function, so as with everything you wish to modify directly from within a function, you need a pointer to it. Since it exists as 'node *head' back in main() or wherever, it's address will be passed as a double pointer to node.
If you were inserting the node at the tail, it would be a different story, but you are inserting at the head and therefore your entry point into the linked list must be modified. Imagine if it weren't.
No.
Are you familar with pass-by-value vs pass-by-pointer? One reason you pass by pointer is if you want to modify the value, as modifying the argument when passed by value has no effect outside of the function.
Now consider what happens if the thing you want to modify is itself a pointer... You have to pass that pointer by pointer! I.e. double-pointer.
My homepage
Advice: Take only as directed - If symptoms persist, please see your debugger
Linus Torvalds: "But it clearly is the only right way. The fact that everybody else does it some other way only means that they are wrong"
It's called a pointer-to-pointer not a "double pointer" which implies pointing to 2 places at once and is impossible in C. Also that name makes it more clear the function in this case
Yes indeed. I should have tried to correct the use of terminology rather than using what term the poster was familar with.
My homepage
Advice: Take only as directed - If symptoms persist, please see your debugger
Linus Torvalds: "But it clearly is the only right way. The fact that everybody else does it some other way only means that they are wrong"
Thanks a lot everyone.
