treap Python实现

# coding=utf-8
# treap(树堆)Python实现

import random


def preorder_tree_walk(node):
    if node:
        print node.key, node.priority
        preorder_tree_walk(node.left)
        preorder_tree_walk(node.right)


def left_rotate(tree, node):
    node_right = node.right
    if not node.p:
        tree.root = node_right
    elif node == node.p.left:
        node.p.left = node_right
    else:
        node.p.right = node_right
    node_right.p = node.p
    node.right = node_right.left
    if node_right.left:
        node_right.left.p = node
    node_right.left = node
    node.p = node_right


def right_rotate(tree, node):
    node_left = node.left
    if not node.p:
        tree.root = node_left
    elif node == node.p.left:
        node.p.left = node_left
    else:
        node.p.right = node_left
    node_left.p = node.p
    node.left = node_left.right
    if node_left.right:
        node_left.right.p = node
    node_left.right = node
    node.p = node_left


class TreapNode(object):
    def __init__(self, key, priority):
        self.key = key
        self.priority = priority
        self.left = None
        self.right = None
        self.p = None


class Treap(object):
    def __init__(self):
        self.root = None

    def find(self, key):
        temp_root = self.root
        while temp_root:
            if key < temp_root.key:
                temp_root = temp_root.left
            elif key > temp_root.key:
                temp_root = temp_root.right
            else:
                return temp_root
        return None

    def insert(self, node):
        temp_root = self.root
        temp_node = None
        while temp_root:
            temp_node = temp_root
            if node.key > temp_node.key:
                temp_root = temp_root.right
            elif node.key < temp_node.key:
                temp_root = temp_root.left
            else:
                raise KeyError, "Error"
        if not temp_node:
            self.root = node
            return
        elif node.key < temp_node.key:
            temp_node.left = node
            node.p = temp_node
        elif node.key > temp_node.key:
            temp_node.right = node
            node.p = temp_node
        self.fixup(temp_node, node)

    def fixup(self, father, child):
        if father:
            if child == father.left and child.priority < father.priority:
                right_rotate(self, father)
            elif child == father.right and child.priority < father.priority:
                left_rotate(self, father)
            self.fixup(father.p, father)

    def delete(self, key):
        node = self.root
        while node:
            if key < node.key:
                node = node.left
            elif key > node.key:
                node = node.right
            else:
                break
        if not node:
            raise KeyError, "Error!"
        flag = 1
        while flag:
            if not node.left:
                if node.right:
                    node.right.p = node.p
                if node.p:
                    if node == node.p.left:
                        node.p.left = node.right
                    else:
                        node.p.right = node.right
                else:
                    self.root = node.right
                flag = 0
            elif not node.right:
                if node.left:
                    node.left.p = node.p
                if node.p:
                    if node == node.p.left:
                        node.p.left = node.left
                    else:
                        node.p.right = node.left
                else:
                    self.root = node.left
                flag = 0
            else:
                # 如果左右子节点均存在 那么把 priority 值大的放到根节点的位置上 网上给的是递归实现 那有没有可能可以递推实现呢？
                # 答案是 有可能的
                if node.left.priority < node.right.priority:
                    right_rotate(self, node)
                else:
                    left_rotate(self, node)


def main():
    number_list = (7, 4, 1, 8, 5, 2, 9, 6, 3)
    tree = Treap()
    for number in number_list:
        priority = random.randint(1, 100)
        node = TreapNode(number, priority)
        tree.insert(node)
    preorder_tree_walk(tree.root)
    print '=========='
    tree.delete(4)
    preorder_tree_walk(tree.root)


if __name__ == '__main__':
    main()

网上一些资料都是递归实现的插入和删除, Python由于没有按引用传递, 所以暂时没有想到好的办法来递归实现, 只好递推实现。
注:暂时没有时间写实现思路，等过一段时间补上
End.