给定一个单链表,随机选择链表的一个节点,并返回相应的节点值。保证每个节点被选的概率一样。
进阶:
如果链表十分大且长度未知,如何解决这个问题?你能否使用常数级空间复杂度实现?
示例:
// 初始化一个单链表 [1,2,3]. ListNode head = new ListNode(1); head.next = new ListNode(2); head.next.next = new ListNode(3); Solution solution = new Solution(head); // getRandom()方法应随机返回1,2,3中的一个,保证每个元素被返回的概率相等。 solution.getRandom();
Related Topics
题解
简单啊,,,随机么。直接存成数组,然后一个
Math.random()
出来了
class Solution2 {
List<ListNode> list;
/** @param head The linked list's head.
Note that the head is guaranteed to be not null, so it contains at least one node. */
public Solution2(ListNode head) {
list = new ArrayList<>();
while (head!=null){
list.add(head);
head = head.next;
}
}
/** Returns a random node's value. */
public int getRandom() {
int random = (int)(Math.random() * list.size());
return list.get(random).val;
}
}
进阶。不用总长度怎么取随机,最近好像遇到好几次了,特地花了点时间去仔细理解了下水塘抽样随机算法。本题题解如下
class Solution {
ListNode head;
Random random;
/** @param head The linked list's head.
Note that the head is guaranteed to be not null, so it contains at least one node. */
public Solution(ListNode head) {
this.head = head;
random = new Random();
}
/** Returns a random node's value. */
public int getRandom() {
ListNode node = head;
int count = 1;
int val = node.val;
while (node!=null){
if (random.nextInt(count) == 1){
val =node.val;
}
count++;
node = node.next;
}
return val;
}
}
发表评论