C# html数据爬取与过滤

1.首先安装第三方HTML数据过滤包　　HtmlAgilityPack

我爬取的网站是一个树洞网站：https://i.jandan.net/treehole，他是一个单体网站，不通过api请求，所以只能根据HTML过滤，他的分页是通过base64加密的

这是获取到的部分数据，这是我们需要的有效数据，他是有固定结构的，我们只要筛选这里面的数据显示出来就好了

以下是所有代码

using HtmlAgilityPack;
using System.Text;
using System.Text.RegularExpressions;

await ReadHtml();

while (true)
{
    Console.WriteLine("请输入页码：");
    var pageNumByNowDate = Console.ReadLine();

    if (!int.TryParse(pageNumByNowDate, out int pageNum))
    {
        Console.WriteLine("页码格式不正确，请重新输入数字");
        continue;
    }

    Console.Clear();

    byte[] buffer = Encoding.UTF8.GetBytes(DateTime.Now.ToString("yyyyMMdd") + "-" + pageNum);

    await ReadHtml("/" + Convert.ToBase64String(buffer));
}

//读取html
async Task ReadHtml(string url = "")
{
    var client = new HttpClient();
    var request = new HttpRequestMessage(HttpMethod.Get, "https://i.jandan.net/treehole" + url);
    request.Headers.Add("Cookie", "PHPSESSID=38e64nulb56bqgl6e27b5sp31l");
    var response = await client.SendAsync(request);
    response.EnsureSuccessStatusCode();
    ClearData(await response.Content.ReadAsStringAsync());
}

//清洗脏数据
void ClearData(string html)
{
    HtmlDocument doc = new HtmlDocument();
    doc.LoadHtml(html);

    var commentTexts = doc.DocumentNode.SelectNodes("//div[@class='commenttext']");

    foreach (var comment in commentTexts)
    {
        var commonText = comment.InnerText.Length > 30 ? string.Join("", comment.InnerText.Select((c, i) => (i > 0 && i % 50 == 0) ? $"{c}\r\n" : c.ToString().Trim().Replace("<br/>", "").Replace("<br />", ""))) : comment.InnerText.Trim().Replace("<br/>", "").Replace("<br />", "");

        var childrens = comment.ParentNode.SelectNodes(".//span[@class='tucao-unlike-container']");

        Console.WriteLine();
        Console.WriteLine();
        Console.WriteLine(commentTexts.IndexOf(comment) + 1 + "、  " + commonText);

        Console.WriteLine();
        Console.WriteLine(string.Join("               ", childrens.Select(c=>c.InnerText.Trim())));

        Console.WriteLine("_____________________________________________________________");
    }
}