如何使用XPath来挖掘(别人的)编码糟糕的HTML

当我执行这段c#代码时…

using System;
using System.Collections.Generic;
using System.ComponentModel;
using System.Data;
using System.Drawing;
using System.Linq;
using System.Text;
using System.Text.RegularExpressions;
using System.Threading.Tasks;
using System.Windows.Forms;
using System.IO;
using System.Net;
using System.Dynamic;
using HtmlAgilityPack;
namespace entropedizer
{
    public partial class Form1 : Form
    {
        public Form1()
        {
            InitializeComponent();
        }
        public String postRequest(string url, string eventTarget)
        {
            // A "pre-request," sent to gather SessionID and POST data parameters for the main request
            HttpWebRequest prequest = (HttpWebRequest)WebRequest.Create("http://www.entropedia.info/Chart.aspx?chart=Chart");
            HttpWebResponse presponse = (HttpWebResponse)prequest.GetResponse();
            Stream pstream = presponse.GetResponseStream();
            StreamReader psr = new StreamReader(pstream);
            string phtml = psr.ReadToEnd();
            Match viewstate = Regex.Match(phtml, "id='"__VIEWSTATE'".+/>");
            Match eventvalidation = Regex.Match(phtml, "id='"__EVENTVALIDATION'".+/>");
            ASCIIEncoding encoding = new ASCIIEncoding();
            string postData = "__EVENTTARGET=" + eventTarget + "&__VIEWSTATE=" + Uri.EscapeDataString(viewstate.ToString().Substring(24, viewstate.Length - 28)) + "&__EVENTVALIDATION=" + Uri.EscapeDataString(eventvalidation.ToString().Substring(30, eventvalidation.Length - 34));
            byte[] data = encoding.GetBytes(postData);
            // The main request, intended to retreive the desired HTML
            HttpWebRequest request = (HttpWebRequest)WebRequest.Create("http://www.entropedia.info/Chart.aspx?chart=Chart");
            request.Method = "POST";
            request.ContentType = "application/x-www-form-urlencoded";
            request.CookieContainer = new CookieContainer();
            Cookie sessionId = new Cookie("ASP.NET_SessionId", Regex.Match(presponse.Headers.ToString(), "ASP.NET_SessionId=.+ d").ToString().Substring(18, Regex.Match(presponse.Headers.ToString(), "ASP.NET_SessionId=.+ d").Length - 21), "/", ".entropedia.info");
            request.CookieContainer.Add(new Uri("http://www.entropedia.info/Chart.aspx?chart=Chart"), sessionId);
            Stream stream = request.GetRequestStream();
            stream.Write(data, 0, data.Length);
            stream.Close();
            HttpWebResponse response = (HttpWebResponse)request.GetResponse();
            stream = response.GetResponseStream();
            StreamReader sr = new StreamReader(stream);
            return sr.ReadToEnd();
        }
        private void Form1_Load(object sender, EventArgs e)
        {
            System.Net.ServicePointManager.Expect100Continue = false;
            HtmlAgilityPack.HtmlDocument hChart = new HtmlAgilityPack.HtmlDocument();
            hChart.LoadHtml(postRequest("http://www.entropedia.info/Chart.aspx?chart=Chart", "ctl00%24ContentPlaceHolder1%24DG1%24ctl19%24ctl05"));
            HtmlNodeCollection chartStrings = hChart.DocumentNode.SelectNodes("/");
            if (chartStrings != null)
            {
                foreach (HtmlNode i in chartStrings)
                {
                    System.IO.File.WriteAllText("C:/Users/Admin/Desktop/WholeDocument.txt", i.OuterHtml);
                }
            }
            else
            {
                MessageBox.Show("Error:  Null item list.");
            }
        }
    }
}

…下面的HTML被写入一个文本文件。

http://pastebin.com/FALerBWR

当我将c#代码中的行更改为HtmlNodeCollection chartStrings = hChart.DocumentNode.SelectNodes("/html/body");时，主体内的400多行HTML将被写入文本文件。

当我将行更改为HtmlNodeCollection chartStrings = hChart.DocumentNode.SelectNodes("/html/body/form");时，只有一行代码(带有其属性的表单开始标记)被写入文本文件。它应该写很多行(大部分文档)。我相信HtmlAgilityPack变得混乱，由于畸形的HTML标签。是否有一种方法可以程序化地解决这个问题?我不想每次运行程序时都手动更正HTML !

HtmlAgilityPack.HtmlDocument hChart = new HtmlAgilityPack.HtmlDocument();
// remove all specific behaviors for the `FORM` element
HtmlAgilityPack.HtmlNode.ElementsFlags.Remove("form");
hChart.LoadHtml(postRequest("http://www.entropedia.info/Chart.aspx?chart=Chart", "ctl00%24ContentPlaceHolder1%24DG1%24ctl19%24ctl05"));
HtmlNodeCollection chartStrings = hChart.DocumentNode.SelectNodes("/");