What is priority queue
The definition in the introduction to the algorithm is shown in the figure below
Algorithm process
Priority queues are implemented using heaps. It just so happens that our above heap sorting can achieve a maximum heap to achieve a priority queue with a little modification:) With the foundation of the previous heap sorting, this implementation makes it easy to understand
MAXIMIUM: as long as Take the element at the top of the heap.
ExtractMax: just take the element at the top of the heap and kick the element out and perform a pile maintenance.
Increase: As long as you keep comparing with the parent node, if it is larger than the parent node, swap position. If it is smaller than the parent node, the position is just right and the loop ends.
insert: After having increase, just add an element to the end of the heap, set its priority to a negative number (representing the minimum priority), and then use the increase method to increase the priority.
Algorithm implementation
class PriorityQueue {
constructor(arr) {
this.queue = []
if (arr && arr instanceof Array && arr.length >= 1) {
this.create(arr)
} else {
throw new Error('非法传入值')
}
}
// 实现一个简单的拷贝
copy(input) {
return JSON.parse(JSON.stringify(input))
}
//获取最高优先级的内容
MAXIMIUM() {
return this.copy(this.queue[0])
}
//插入一个新的元素
insert(target) {
if (!target || !target.priority) return false
let temp = target.priority
target.priority = -1
this.queue.push(target)
this.increase(this.queue.length-1,temp)
}
//提升一个元素的优先级
increase(index, newPriority) {
if (newPriority <= this.queue[index].priority) return
this.queue[index].priority = newPriority
const findParent = (index) => {
if (index == 0) {
return null
}
return index % 2 == 0 ? (index - 2) / 2 : (index - 1) / 2
}
let parent = findParent(index)
while (parent != null && this.queue[parent].priority < this.queue[index].priority) {
let box = this.queue[index]
this.queue[index] = this.queue[parent]
this.queue[parent] = box
index = parent
parent = findParent(parent)
}
}
//获取最高优先级的内容 然后踢出队列
extractMax() {
let max = this.copy(this.queue[0])
this.queue.shift()
if (this.queue.length != 1) {
this.MaxHeap(this.queue, 0, this.queue.length)
}
return max
}
//提供一个初始化方法 建堆
create(arr) {
this.queue = this.copy(arr)
this.BuildMaxHeap(this.queue, this.queue.length)
}
MaxHeap(arr, i, length) {
var largest = null;
var node = arr[i]; //保存当前节点
var left = i * 2 + 1; //定位节点左
var right = i * 2 + 2; //定位节点右
//判断当前有这个节点 （这里会存在当前这个的子节点不存在的情况）处理一下边界情况
if (left < length) {
arr[left] && node.priority < arr[left].priority ? largest = left : largest = i
} else {
largest = i;
}
if (right < length) {
arr[right] && arr[largest].priority < arr[right].priority ? largest = right : null
}
//如果不是i是最大节点 以node作为辅助节点 交换位置
if (largest != i) {
arr[i] = arr[largest];
arr[largest] = node;
this.MaxHeap(arr, largest, length);
}
}
//建立最大堆
BuildMaxHeap(arr, len) {
if (len % 2 != 0) {
len = len + 1;
}
for (let i = len / 2; i >= 0; i--) {
this.MaxHeap(arr, i, len)
}
}
}
let pq = new PriorityQueue([
{
priority: 1, todo: '吃饭' },
{
priority: 2, todo: '洗澡' },
{
priority: 3, todo: '睡觉' },
{
priority: 4, todo: '玩手机' },
{
priority: 5, todo: '打篮球' },
{
priority: 6, todo: '打乒乓' },
{
priority: 7, todo: '喝水' }
])
console.log(pq.queue)
pq.increase(6, 8)
console.log(pq.queue)
pq.insert({
priority: 9, todo: '我是新加的'})
console.log(pq.queue)