Design and implement a data structure for Least Recently Used (LRU) cache. It should support the following operations: get and set.node
get(key) - Get the value (will always be positive) of the key if the key exists in the cache, otherwise return -1.數據結構
set(key, value) - Set or insert the value if the key is not already present. When the cache reached its capacity, it should invalidate the least recently used item before inserting a new item.this
這題須要咱們設計一個cache, 有get和set兩個操做。由於是cache, 全部兩個操做的時間複雜度都必須是O(1)。
get(key) -- O(1) 很明顯,咱們須要用一個hashmap來實現O(1)的操做。
set(key, value) -- O(1) 這裏有兩種狀況,key沒出現過,就直接加在head。這裏出現一個關鍵詞head。
由於使用線性的數據結構,而且表示操做的前後順序,這樣的結構就是鏈表。是單鏈表仍是雙鏈表?下面咱們模擬一下:
capacity = 3
set(1, 100)
set(2, 200)
set(3, 300)
get(2)
若是是單鏈表,簡單表示以下:
1 -> 2 -> 3 -> null
咱們能夠獲得2並放在頭部。可是這裏用的單鏈表,咱們沒法知道2的前面是什麼,2前面的全部點都會脫離總體。因此須要一個雙鏈表。
1 <=> 3 <=> 2 <=> null
咱們繼續操做,set(4, 400),發現已經達到LRU的容量,須要移除,這時候發現咱們須要一個尾部來告訴咱們須要移除哪一個點。
咱們發現,不管是get(key)仍是set(key, value)都有兩個簡單操做組成,從鏈表中移除,放到鏈表頭部。
能夠定義兩個helper function: remove(node), setHead(node)。debug
代碼以下,帶註釋:設計
public class LRUCache { class Node{ int key; int value; Node pre; // point to tail direction Node next; // point to head direction public Node(int key, int value){ this.key = key; this.value = value; } } int capacity; Map<Integer, Node> map = new HashMap<>(); Node tail = null; Node head = null; public LRUCache(int capacity) { this.capacity = capacity; } public int get(int key) { if(map.containsKey(key)){ // remove from LRU and put it to head of LRU Node n = map.get(key); remove(n); setHead(n); return n.value; } return -1; } public void set(int key, int value) { if(map.containsKey(key)){ // change node value, remove from LRU and put it to head of LRU Node old = map.get(key); old.value = value; remove(old); setHead(old); } else { Node newNode = new Node(key, value); if(capacity == map.size()){ // remove the tail map.remove(tail.key); remove(tail); } setHead(newNode); // set newNode to head map.put(key, newNode); } } public void remove(Node n){ if(n.pre != null) { // change pre node connection n.pre.next = n.next; } else { // check if it is the tail tail = n.next; } if(n.next != null) { // change next node connection n.next.pre = n.pre; } else { // check if it is the head head = n.pre; } } public void setHead(Node n){ n.pre = head; n.next = null; if(head != null) { // check head exist or Not ? head.next = n; } head = n; if(tail == null){ // empty LRU, intitailize tail node tail = head; } } }
使用dummyEnd 和dummyHead能夠簡化代碼。code
public class LRUCache { int capacity; Map<Integer, Node> map; Node dummyEnd; Node dummyHead; int count; public LRUCache(int capacity) { this.capacity = capacity; this.count = 0; map = new HashMap<Integer, Node>(); dummyEnd = new Node(0,0); dummyHead = new Node(0,0); dummyEnd.next = dummyHead; dummyHead.pre = dummyEnd; } public int get(int key) { Node node = map.get(key); if(node == null) { return -1; } else { remove(node); putToHead(node); return node.val; } } public void put(int key, int value) { Node oldNode = map.get(key); if(oldNode == null) { ++count; Node newNode = new Node(key, value); map.put(key, newNode); putToHead(newNode); if(count > capacity){ // 從LRU移除 // 第一次在這裏debug了很久,要先取出nextNode, 否則map裏remove的就是錯誤的點,即dummy.next.next。 Node nextNode = dummyEnd.next; remove(nextNode); // 從map移除 map.remove(nextNode.key); --count; } } else { // 改變值,先移除,再放入頭部 oldNode.val = value; remove(oldNode); putToHead(oldNode); } } public void putToHead(Node node){ // 加到頭和前一個點的中間 Node preNode = dummyHead.pre; preNode.next = node; node.pre = preNode; dummyHead.pre = node; node.next = dummyHead; } public void remove(Node node){ // 移除。 node.next.pre = node.pre; node.pre.next = node.next; // node 若是從尾部移除,將不會指向任何點。 node.pre = null; node.next = null; } class Node{ int key, val; Node pre, next; public Node(int key, int val){ this.key = key; this.val = val; } } } /** * Your LRUCache object will be instantiated and called as such: * LRUCache obj = new LRUCache(capacity); * int param_1 = obj.get(key); * obj.put(key,value); */