I am trying to use a double void
pointer but I am a little bit confused about the usage.
I have a struct
that contains a void **
array.
struct Thing{
void ** array;
};
struct Thing * c = malloc (sizeof(struct Thing));
c->array = malloc( 10 * sizeof(void *) );
So If I want to assign a different object to each pointer and try to retrieve the value
// Option 1
*(c->array + index) = (void *) some_object_ptr;
// Option 2
c->array[index] = (void *) some_object_ptr;
then, I have another function that gives (void *) item
that points to each cell, not the some_object_ptr
.
If I want to retrieve the value which pointed to by some_object_ptr
,
should I do
function return type is 'void *' and takes argument 'void *'
// Option 3
return (void**) item
// Option 4
return *((void**)item)?
the weird thing is that when I used array the array subscript method I couldn't use option 4, only option 3; and when I used *(c->array + index)
I could only use opt.4. and not opt.3. ..
Can anyone please tell me about this? If I am making any invalid assumptions, then could you please correct me?
A void **
is just a pointer to a pointer to memory with an unspecified type. You can only dereference it once (since you can't dereference a void *
). However, apart from that, it is basically like any other pointer type. If it helps you, think of it the same way as you would with int *
.
So, in your specific situation, we have:
void** array;
int arrayLen = 10;
array = (void**)malloc(arrayLen * sizeof(void*));
some_type_t* some_object_ptr;
// The following two assignment are equivalent since in C,
// array[index] <=> *(array + index)
array[index] = (void*)some_object_ptr;
*(array + index) = (void*)some_object_ptr;
Then array
is a pointer to the whole array, while *array
is a pointer to the first element, since it is equivalent to array[0]
.