题解 | #设计LFU缓存结构#

import java.util.*;
public class Solution {
    //设置节点结构
    static class Node{ 
        int freq;
        int key;
        int val;
        //初始化
        public Node(int freq, int key, int val) {
            this.freq = freq;
            this.key = key;
            this.val = val;
        }
    }
    //频率到双向链表的哈希表
    private Map<Integer, LinkedList<Node> > freq_mp = new HashMap<>();
    //key到节点的哈希表
    private Map<Integer, Node> mp = new HashMap<>();
    //记录缓存剩余容量
    private int size = 0; 
    //记录当前最小频次
    private int min_freq = 0;
    
    public int[] LFU (int[][] operators, int k) {
        //构建初始化连接
        //链表剩余大小
        this.size = k;
        //获取操作数
        int len = (int)Arrays.stream(operators).filter(x -> x[0] == 2).count();
        int[] res = new int[len];
        //遍历所有操作
        for(int i = 0, j = 0; i < operators.length; i++){
            if(operators[i][0] == 1)
                //set操作
                set(operators[i][1], operators[i][2]);
            else
                //get操作
                res[j++] = get(operators[i][1]);
        }
        return res;
    }
    
    //调用函数时更新频率或者val值
    private void update(Node node, int key, int value) { 
        //找到频率
        int freq = node.freq;
        //原频率中删除该节点
        freq_mp.get(freq).remove(node); 
        //哈希表中该频率已无节点，直接删除
        if(freq_mp.get(freq).isEmpty()){ 
            freq_mp.remove(freq);
            //若当前频率为最小，最小频率加1
            if(min_freq == freq) 
                min_freq++;
        }
        if(!freq_mp.containsKey(freq + 1))
            freq_mp.put(freq + 1, new LinkedList<Node>());
        //插入频率加一的双向链表表头，链表中对应：freq key value
        freq_mp.get(freq + 1).addFirst(new Node(freq + 1, key, value)); 
        mp.put(key, freq_mp.get(freq + 1).getFirst());
    }
    
    //set操作函数
    private void set(int key, int value) {
        //在哈希表中找到key值
        if(mp.containsKey(key)) 
            //若是哈希表中有，则更新值与频率
            update(mp.get(key), key, value);
        else{ 
            //哈希表中没有，即链表中没有
            if(size == 0){
                //满容量取频率最低且最早的删掉
                int oldkey = freq_mp.get(min_freq).getLast().key; 
                //频率哈希表的链表中删除
                freq_mp.get(min_freq).removeLast(); 
                if(freq_mp.get(min_freq).isEmpty()) 
                    freq_mp.remove(min_freq); 
                //链表哈希表中删除
                mp.remove(oldkey); 
            }
            //若有空闲则直接加入，容量减1
            else 
                size--; 
            //最小频率置为1
            min_freq = 1; 
            //在频率为1的双向链表表头插入该键
            if(!freq_mp.containsKey(1))
                freq_mp.put(1, new LinkedList<Node>());
            freq_mp.get(1).addFirst(new Node(1, key, value)); 
            //哈希表key值指向链表中该位置
            mp.put(key, freq_mp.get(1).getFirst()); 
        }
    }
    
    //get操作函数
    private int get(int key) {
        int res = -1;
        //查找哈希表
        if(mp.containsKey(key)){ 
            Node node = mp.get(key);
            //根据哈希表直接获取值
            res = node.val;
            //更新频率 
            update(node, key, res); 
        }
        return res;
    }
}