先演示一下效果,再展示代码逻辑。
中间几次调用过程省略。。。
暂时只用到了下面四个项目
1.产品展示页面中第一次通过接口去获取数据库的列表数据
////// 获取指定的商品目录 /// /// /// /// /// [HttpGet] [Route("items")] [ProducesResponseType(typeof(PaginatedViewModel), StatusCodes.Status200OK)] [ProducesResponseType(typeof(IEnumerable), StatusCodes.Status200OK)] [ProducesResponseType(StatusCodes.Status400BadRequest)] public async Task Catalogs([FromQuery] int pageSize = 10, [FromQuery] int pageIndex = 0, string ids = null) { if (!string.IsNullOrEmpty(ids)) { var items = await GetItemByIds(ids); if (!items.Any()) { return BadRequest("ids value invalid. Must be comma-separated list of numbers"); } return Ok(items); } var totalItems = await _catalogContext.Catalogs .LongCountAsync(); var itemsOnPage = await _catalogContext.Catalogs .OrderBy(c => c.Name) .Skip(pageSize * pageIndex) .Take(pageSize) .ToListAsync(); var result = itemsOnPage.Select(x => new ProductDto(x.Id.ToString(), x.Name, x.Price.ToString(), x.Stock.ToString(), x.ImgPath)); var model = new PaginatedViewModel(pageIndex, pageSize, totalItems, result); return Ok(model); }
2.在前端页面会把当前页面的产品列表id都发送到websocket中去
function updateAndSendProductIds(ids) { productIds = ids; // Check if the WebSocket is open if (socket.readyState === WebSocket.OPEN) { // Send the list of product IDs through the WebSocket connection socket.send(JSON.stringify(productIds)); } } function fetchData() { const apiUrl = baseUrl + `/Catalog/items?pageSize=${pageSize}&pageIndex=${currentPage}`; axios.get(apiUrl) .then(response => { const data = response.data.data; displayProducts(baseUrl, data); const newProductIds = data.map(product => product.Id); 服务器托管网 // Check if the WebSocket is open updateAndSendProductIds(newProductIds); // 从响应中获取总页数 const totalPages = Math.ceil(response.data.count / pageSize); displayPagination(totalPages); // 更新当前页数的显示 const currentPageElement = document.getElementById('currentPage'); currentPageElement.textContent = `当前页数: ${currentPage + 1} / 总页数: ${totalPages}`; }) .catch(error => { console.error('获取数据失败:', error); }); }
3.websocket拿到了id数据可以精确的把当前页面的产品都查出来再推送给product.html页面,通过下面的ReceiveAsync方法获取html发送的数据,再通过timer定时器每秒钟Send方法实时的往页面发送获取到的数据,当然这个是不断的去从redis中去查的。
using System.Net.WebSockets; using System.Threading.Tasks; using System; using WsServer.Handler; using WsServer.Manager; using StackExchange.Redis; using Microsoft.Extensions.Configuration; using System.Collections.Generic; using Catalogs.Domain.Catalogs; using Catalogs.Domain.Dtos; using System.Net.Sockets; namespace WebScoket.Server.Services { ////// 实时推送产品主要是最新的库存,其他信息也会更新 /// public class ProductListHandler : WebSocketHandler { private System.Threading.Timer _timer; private readonly IDatabase _redisDb; //展示列表推送 private string productIdsStr; public ProductListHandler(WebSocketConnectionManager webSocketConnectionManager,IConfiguration configuration) : base(webSocketConnectionManager) { ConnectionMultiplexer redis = ConnectionMultiplexer.Connect(configuration["DistributedRedis:ConnectionString"] ?? throw new Exception("$未能获取distributedredis连接字符串")); _redisDb = redis.GetDatabase(); _timer = new System.Threading.Timer(Send, null, TimeSpan.Zero, TimeSpan.FromSeconds(1)); } private void Send(object state) { // 获取当前时间并发送给所有连接的客户端 if (productIdsStr != null) { string[] productIds = System.Text.Json.JsonSerializer.Deserialize(productIdsStr); string hashKeyToRetrieve = "products"; List products = new List(); foreach (var productId in productIds) { if(productId == "null") { continue; } string retrievedProductValue = _redisDb.HashGet(hashKeyToRetrieve, productId); if (!string.IsNullOrEmpty(retrievedProductValue)) { //反序列化和构造函数冲突,改造了一下Catalog Catalog catalog = System.Text.Json.JsonSerializer.Deserialize(retrievedProductValue); products.Add(new ProductDto(catalog.Id.ToString(), catalog.Name, catalog.Price.ToString(), catalog.Stock.ToString(), catalog.ImgPath)); } } if (products.Count > 0) { SendMessageToAllAsync(System.Text.Json.JsonSerializer.Serialize(products)).Wait(); 服务器托管网 } else { SendMessageToAllAsync("NoProduct").Wait(); } } } public override async Task ReceiveAsync(WebSocket socket, WebSocketReceiveResult result, byte[] buffer) { //每次页面有刷新就会拿到展示的id列表 productIdsStr = System.Text.Encoding.UTF8.GetString(buffer, 0, result.Count); } } }
4.html页面就可以拿到最新数据再去绑定到页面
socket.addEventListener('message', (event) => { if (event.data == "NoProduct") { clearProductList(); } // Handle the received product data and update the product list const productData = JSON.parse(event.data); // Update the product list with the received data (call your displayProducts function) displayProducts(baseUrl, productData); });
整个流程就这么简单,但是这里需要保持数据库和redis的数据实时同步,否则页面展示的就不是最新的数据就没意义了。
再回到Catalog.Service服务中。
private async Task DeleteCache()
{
//await _redisDb.HashDeleteAsync("products",id); //没必要了
await _channel.Writer.WriteAsync("delete_catalog_fromredis");
}
再做更新、新增、删除等动作的时候就调用一下DeleteCache方法,往后台服务发送一个channel,当后台收到后就做redis删除并且从初始化sqlserver到redis列表同步的操作
using System.Reflection; using System.Threading.Channels; using Catalogs.Infrastructure.Database; using Microsoft.EntityFrameworkCore; using Microsoft.Extensions.Hosting; using Microsoft.Extensions.Logging; using StackExchange.Redis; namespace Catalogs.WebApi.BackgroudServices { /// summary> /// 记得任何删除了或者购买了产品后需要删除改产品的键 /// summary> public class InitProductListToRedisService : BackgroundService { private readonly IServiceScopeFactory _serviceScopeFactory; private readonly IDatabase _redisDb; private readonly Channelstring> _channel; private readonly ILogger _logger; public InitProductListToRedisService(IServiceScopeFactory serviceScopeFactory, IConfiguration configuration, Channelstring> channel, ILoggerInitProductListToRedisService> logger) { _serviceScopeFactory = serviceScopeFactory; ConnectionMultiplexer redis = ConnectionMultiplexer.Connect(configuration["DistributedRedis:ConnectionString"] ?? throw new Exception("$未能获取distributedredis连接字符串")); _redisDb = redis.GetDatabase(); _channel = channel; _logger = logger; } protected override async Task ExecuteAsync(CancellationToken stoppingToken) { await Init(); while (!_channel.Reader.Completion.IsCompleted) { var msg = await _channel.Reader.ReadAsync(); if(msg == "delete_catalog_fromredis") { await Init(); } } } private async Task Init() { using var scope = _serviceScopeFactory.CreateScope(); try { CatalogContext _context = scope.ServiceProvider.GetRequiredServiceCatalogContext>(); string hashKey = "products"; var products = await _context.Catalogs.ToListAsync(); await _redisDb.KeyDeleteAsync(hashKey); foreach (var product in products) { string productField = product.Id.ToString(); string productValue = System.Text.Json.JsonSerializer.Serialize(product); _redisDb.HashSet(hashKey, new HashEntry[] { new HashEntry(productField, productValue) }); } _logger.LogInformation($"ProductList is over stored in Redis Hash."); } catch(Exception ex) { _logger.LogError($"ProductLis stored in Redis Hash error."); } } } }
这里还有优化的空间可以只针对怕products的hashset的某个id去更新、删除、新增一条数据。
示例代码:
liuzhixin405/efcore-template (github.com)
服务器托管,北京服务器托管,服务器租用 http://www.fwqtg.net
机房租用,北京机房租用,IDC机房托管, http://www.fwqtg.net
相关推荐: Spring5_AOP_jdk 和 cglib 在 Spring 中的统一、通知调用
jdk 和 cglib 在 Spring 中的统一 Spring 中对切点、通知、切面的抽象如下 切点:接口 Pointcut,典型实服务器托管网现 AspectJExpressionPointcut 通知:典型接口为 MethodInterceptor 代表…