2. it
  3. 红黑树Python实现

红黑树Python实现

mac2022-06-30  92
# coding=utf-8 # 红黑树Python实现 # 颜色常量 RED = 0 BLACK = 1 def left_rotate(tree, node): if not node.right: return False node_right = node.right node_right.p = node.p if not node.p: tree.root = node_right elif node == node.p.left: node.p.left = node_right else: node.p.right = node_right if node_right.left: node_right.left.p = node node.right = node_right.left node.p = node_right node_right.left = node def right_rotate(tree, node): if not node.left: return False node_left = node.left node_left.p = node.p if not node.p: tree.root = node_left elif node == node.p.left: node.p.left = node_left elif node == node.p.right: node.p.right = node_left if node_left.right: node_left.right.p = node node.left = node_left.right node.p = node_left node_left.right = node def transplant(tree, node_u, node_v): """ 用 v 替换 u :param tree: 树的根节点 :param node_u: 将被替换的节点 :param node_v: 替换后的节点 :return: None """ if not node_u.p: tree.root = node_v elif node_u == node_u.p.left: node_u.p.left = node_v elif node_u == node_u.p.right: node_u.p.right = node_v # 加一下为空的判断 if node_v: node_v.p = node_u.p def tree_maximum(node): """ 找到以 node 节点为根节点的树的最大值节点 并返回 :param node: 以该节点为根节点的树 :return: 最大值节点 """ temp_node = node while temp_node.right: temp_node = temp_node.right return temp_node def tree_minimum(node): """ 找到以 node 节点为根节点的树的最小值节点 并返回 :param node: 以该节点为根节点的树 :return: 最小值节点 """ temp_node = node while temp_node.left: temp_node = temp_node.left return temp_node def preorder_tree_walk(node): if node: print (node.value, node.color) preorder_tree_walk(node.left) preorder_tree_walk(node.right) class RedBlackTreeNode(object): def __init__(self, value): self.value = value self.left = None self.right = None self.p = None self.color = RED class RedBlackTree(object): def __init__(self): self.root = None def insert(self, node): # 找到最接近的节点 temp_root = self.root temp_node = None while temp_root: temp_node = temp_root if node.value == temp_node.value: return False elif node.value > temp_node.value: temp_root = temp_root.right else: temp_root = temp_root.left # 在相应位置插入节点 if not temp_node: self.root = node node.color = BLACK elif node.value < temp_node.value: temp_node.left = node node.p = temp_node else: temp_node.right = node node.p = temp_node # 调整树 self.insert_fixup(node) def insert_fixup(self, node): if node.value == self.root.value: return # 为什么是这个终止条件？ # 因为如果不是这个终止条件那就不需要调整 while node.p and node.p.color == RED: # 只要进入循环则必有祖父节点 否则父节点为根节点 根节点颜色为黑色 不会进入循环 if node.p == node.p.p.left: node_uncle = node.p.p.right # 1. 没有叔叔节点 若此节点为父节点的右子 则先左旋再右旋 否则直接右旋 # 2. 有叔叔节点 叔叔节点颜色为黑色 # 3. 有叔叔节点 叔叔节点颜色为红色 父节点颜色置黑 叔叔节点颜色置黑 祖父节点颜色置红 continue # 注: 1 2 情况可以合为一起讨论 父节点为祖父节点右子情况相同 只需要改指针指向即可 if node_uncle and node_uncle.color == RED: node.p.color = BLACK node_uncle.color = BLACK node.p.p.color = RED node = node.p.p continue elif node == node.p.right: left_rotate(self, node.p) node = node.left node.p.color = BLACK node.p.p.color = RED right_rotate(self, node.p.p) return elif node.p == node.p.p.right: node_uncle = node.p.p.left if node_uncle and node_uncle.color == RED: node.p.color = BLACK node_uncle.color = BLACK node.p.p.color = RED node = node.p.p continue elif node == node.p.left: right_rotate(self, node) node = node.right node.p.color = BLACK node.p.p.color = RED left_rotate(self, node.p.p) return # 最后记得把根节点的颜色改为黑色 保证红黑树特性 self.root.color = BLACK def delete(self, node): # 找到以该节点为根节点的右子树的最小节点 node_color = node.color if not node.left: temp_node = node.right transplant(self, node, node.right) elif not node.right: temp_node = node.left transplant(self, node, node.left) else: # 最麻烦的一种情况 既有左子 又有右子 找到右子中最小的做替换 类似于二分查找树的删除 node_min = tree_minimum(node.right) node_color = node_min.color temp_node = node_min.right if node_min.p != node: transplant(self, node_min, node_min.right) node_min.right = node.right node_min.right.p = node_min transplant(self, node, node_min) node_min.left = node.left node_min.left.p = node_min node_min.color = node.color # 当删除的节点的颜色为黑色时 需要调整红黑树 if node_color == BLACK: self.delete_fixup(temp_node) def delete_fixup(self, node): # 实现过程还需要理解 比如为什么要删除 为什么是那几种情况 while node != self.root and node.color == BLACK: if node == node.p.left: node_brother = node.p.right if node_brother.color == RED: node_brother.color = BLACK node.p.color = RED left_rotate(self, node.p) node_brother = node.p.right if (not node_brother.left or node_brother.left.color == BLACK) and \ (not node_brother.right or node_brother.right.color == BLACK): node_brother.color = RED node = node.p else: if not node_brother.right or node_brother.right.color == BLACK: node_brother.color = RED node_brother.left.color = BLACK right_rotate(self, node_brother) node_brother = node.p.right node_brother.color = node.p.color node.p.color = BLACK node_brother.right.color = BLACK left_rotate(self, node.p) node = self.root break else: node_brother = node.p.left if node_brother.color == RED: node_brother.color = BLACK node.p.color = RED left_rotate(self, node.p) node_brother = node.p.right if (not node_brother.left or node_brother.left.color == BLACK) and \ (not node_brother.right or node_brother.right.color == BLACK): node_brother.color = RED node = node.p else: if not node_brother.left or node_brother.left.color == BLACK: node_brother.color = RED node_brother.right.color = BLACK left_rotate(self, node_brother) node_brother = node.p.left node_brother.color = node.p.color node.p.color = BLACK node_brother.left.color = BLACK right_rotate(self, node.p) node = self.root break node.color = BLACK def main(): number_list = (7, 4, 1, 8, 5, 2, 9, 6, 3) tree = RedBlackTree() for number in number_list: node = RedBlackTreeNode(number) tree.insert(node) del node preorder_tree_walk(tree.root) tree.delete(tree.root) preorder_tree_walk(tree.root) if __name__ == '__main__': main() 教材: 算法导论(第三版)

转载于:https://www.cnblogs.com/wuditju/p/5992129.html

最新回复(0)