迭代二叉搜索树插入C语言

Question

This seems to be a simple question, and I have looked at another thread and it seems I am doing the same thing, however without results.

This is my code to iteratively insert into a binary search tree, along with the structure and how to create a new node:

typedef struct node {
    int data;
    struct node *left;
    struct node *right;
} node;

node *create_node(int data) {
    node *n = calloc(1, sizeof(node));

    n->data = data;
    return n;
}

node *BST_insert_iterative(node *root, int data) {
    node *temp = root;

    if (root == NULL)
        return create_node(data);

    while (temp != NULL) {
        if (data > temp->data)
            temp = temp->right;
        else
            temp = temp->left;
    }

    temp = create_node(data);

    return root;
}

When I print the tree I only receive the first node:

Inserting 8  into BST...
Inserting 50     into BST...
Inserting 74     into BST...
Inserting 59     into BST...
Inserting 31     into BST...
Inserting 73     into BST...
Inserting 45     into BST...
Inserting 79     into BST...
Inserting 24     into BST...
Inserting 10     into BST...
In order traversal
8
Height of tree: 0

However, using the recursive insert function:

node *BST_insert(node *root, int data) {
    if (root == NULL)
        return create_node(data);

    if (root->data >= data)
        root->left = BST_insert(root->left, data);

    else (root->data < data)
        root->right = BST_insert(root->right, data);

    return root;
}

it works just fine and I get:

Inserting 8  into BST...
Inserting 50     into BST...
Inserting 74     into BST...
Inserting 59     into BST...
Inserting 31     into BST...
Inserting 73     into BST...
Inserting 45     into BST...
Inserting 79     into BST...
Inserting 24     into BST...
Inserting 10     into BST...
In order traversal
8
10
24
31
45
50
59
73
74
79
Height of tree: 4 

Answer 1

In your insert function, you do not store the new node into the tree. You merely store in to a local variable temp and always return root .

You must keep a pointer to the link to be updated so the new node is inserted into the tree, or at the root of the tree. Here is a modified version:

typedef struct node {
    int data;
    struct node *left;
    struct node *right;
} node;

node *create_node(int data) {
    node *n = calloc(1, sizeof(node));
    n->data = data;
    n->left = n->right = NULL;
    return n;
}

node *BST_insert_iterative(node *root, int data) {
    node **pp = &root;

    while (*pp != NULL) {
        if (data > (*pp)->data)
            pp = &(*pp)->right;
        else
            pp = &(*pp)->left;
    }
    *pp = create_node(data);
    return root;
}

Notes:

• There is no special case for an empty tree.
• Be aware that this approach will not be sufficient to keep the tree balanced.

Note also that your recursive function has a syntax error in what looks like a redundant test. You should simplify it this way:

node *BST_insert(node *root, int data) {
    if (root == NULL)
        return create_node(data);

    if (root->data >= data) {
        root->left = BST_insert(root->left, data);
    } else {
        root->right = BST_insert(root->right, data);
    }
    return root;
}

Answer 2

Chqrlie's answer worked, but I wanted to do it without using a double pointer.

I was able to do it after realizing my root was not connected to the newly created node.

Here is that solution:

node *BST_insert_iterative(node *root, int data)
{
    node *temp = root;
    int condition = 1;

    if (root == NULL)
        return create_node(data);

    while (condition)
    {
        if (data > temp->data)
        {
            if (temp->right == NULL)
            {
                temp->right = create_node(data);
                condition = 0;
            }
            else
                temp = temp->right;
        }
        else
        {
            if (temp->left == NULL)
            {
                temp->left = create_node(data);
                condition = 0;
            }
            else
                temp = temp->left;
        }

    }

    return root;
}