编码二叉搜索树

Question

According to my coding class, the left child is 2 * i and the right child is 2 * i + 1 .

With this in mind, I coded the following:

void max_heap( int key ){
    int leftkey = ( 2 * key ) + 1;
    int rigtkey = ( 2 * key ) + 2;
    printf( "key is: %d left child index is: %d right child index is: %d \n",
        key, leftkey, rigtkey );
    printf( "value at key is: %d left child is: %d right child is: %d \n",
        heap[key], heap[leftkey], heap[rigtkey] );

    if ( key >= 0 ){ // My base case?
        if ( leftkey < size && leftkey != size ){
            //Makes sure I don't go out of bounds

            if ( heap[leftkey] > heap[key] ){
                swap( &heap[leftkey], &heap[key] );

            }
        }
        else if ( rigtkey < size && rigtkey != size ){
            // Makes sure I don't go out of bounds

            if ( heap[rigtkey] > heap[key] ){
                swap( &heap[rigtkey], &heap[key] );

            }
        }
        max_heap( key-- );

    }
}

I call the method with max_heap(size/2) , with size being the number of entries in the array.

When I run the program and enter:

1 2 3 4 5 6

The result is:

key is: 3 left child index is: 7 right child index is: 8 
value at key is: 4 left child is: 0 right child is: 0 

key is: 3 left child index is: 7 right child index is: 8 
value at key is: 4 left child is: 0 right child is: 0 

key is: 3 left child index is: 7 right child index is: 8 
value at key is: 4 left child is: 0 right child is: 0 

key is: 3 left child index is: 7 right child index is: 8 
value at key is: 4 left child is: 0 right child is: 0 

...

It loops forever like that and no swapping is ever done.

Answer 1

The problem is with how you're recursing. You're recursing with max_heap(key--); which effectively means "first call max_heap(key); and then set key = key - 1; ". This gives you an endless loop since each call to your function is made with the same key that was originally provided. You will get better results if you pre-decrement ( max_heap(--key); ).

That being said, this is not a good problem to solve with recursion. A while loop would be much better.