C# 获取网页上指定的元素(.net Framework4.0 兼容版本)

发布时间 2023-08-16 23:59:20作者: 孤幽影暗
using System;
using System.IO;
using System.Net;
using System.Net.Security;
using System.Security.Cryptography.X509Certificates;
using System.Text;
using HtmlAgilityPack;
using Newtonsoft.Json;

class Program
{
    static void Main(string[] args)
    {
        var url = "https://go-upc.com/search?q=776545980144"; // 替换为你要获取数据的网页地址
        var htmlContent = DownloadHtml(url);

        var htmlDocument = new HtmlDocument();
        htmlDocument.LoadHtml(htmlContent);

        var productNameNode = htmlDocument.DocumentNode.SelectSingleNode("//h1[@class='product-name']");
        var eanNode = htmlDocument.DocumentNode.SelectSingleNode("//td[@class='metadata-label'][contains(text(), 'EAN')]/following-sibling::td");
        var upcNode = htmlDocument.DocumentNode.SelectSingleNode("//td[@class='metadata-label'][contains(text(), 'UPC')]/following-sibling::td");
        var descriptionNode = htmlDocument.DocumentNode.SelectSingleNode("//h2[contains(text(), 'Description')]/following-sibling::span");

        if (productNameNode != null && eanNode != null && upcNode != null && descriptionNode != null)
        {
            var productName = productNameNode.InnerText.Trim();
            var ean = eanNode.InnerText.Trim();
            var upc = upcNode.InnerText.Trim();
            var description = descriptionNode.InnerText.Trim();

            var jsonData = new
            {
                ProductName = productName,
                EAN = ean,
                UPC = upc,
                Description = description
            };

            var jsonResult = JsonConvert.SerializeObject(jsonData);

            Console.WriteLine(jsonResult); // 返回JSON数据给前端
            Console.ReadLine();
        }
    }
    private static bool CheckValidationResult(object sender, X509Certificate certificate, X509Chain chain, SslPolicyErrors errors)
    {
        return true;
    }
    static string DownloadHtml(string url)
    {
        string htmlContent = null;
        ServicePointManager.ServerCertificateValidationCallback = new RemoteCertificateValidationCallback(CheckValidationResult);
        // 重點是修改這行
        ServicePointManager.SecurityProtocol = (SecurityProtocolType)3072; // SecurityProtocolType.Tls1.2
        using (WebClient client = new WebClient())
        {
            client.Encoding = Encoding.UTF8;
            htmlContent = client.DownloadString(url);
        }

        return htmlContent;
    }
}