本身作Fiddler，實現http網站的抓取

---恢復內容開始---

因爲工做須要，須要作一個小程序，實現網站監控，當瀏覽器瀏覽到目標網站的時候，實現爬取數據。因爲https存在證書驗證，須要別的方式來，暫時還沒研究，但必須能正常訪問。

官方的Demo，我在這個基礎上進行了修改。http://fiddler.wikidot.com/fiddlercore-demo

這邊我來介紹下個人Demo。

首先要去下載FiddlerCore4.Dll，百度上不少。

先上代碼：

主界面代碼

using System;
using System.Collections.Generic;
using System.IO;
using System.Reflection;
using System.Threading;
using Fiddler;
using System.Net;
using System.Text.RegularExpressions;
using System.Text;
using System.Diagnostics;
using System.Runtime.InteropServices;

namespace FiddlerDemo
{
    class Program
    {
        public static Proxy oProxy;
        //static string sSecureEndpointHostname = "cd.chnai.com";
        //static int iSecureEndpointPort = 7777;
       
        #region 控制檯異常關閉時，程序不影響電腦
        public delegate bool ControlCtrlDelegate(int CtrlType);
        [DllImport("kernel32.dll")]
        private static extern bool SetConsoleCtrlHandler(ControlCtrlDelegate HandlerRoutine, bool Add);
        private static ControlCtrlDelegate cancelHandler = new ControlCtrlDelegate(HandlerRoutine);

        public static bool HandlerRoutine(int CtrlType)
        {
            switch (CtrlType)
            {
                case 0:
                    DoQuit(); //Ctrl+C關閉 
                    break;
                case 2:
                    //Console.WriteLine("2工具被強制關閉");//按控制檯關閉按鈕關閉 
                    DoQuit();
                    break;
            }
            return false;
        }
        #endregion
        static void Main(string[] args)
        {
            SetConsoleCtrlHandler(cancelHandler, true);
            FiddlerHelp fh = new FiddlerHelp();
            fh.StartSession();
        }
        static void Console_CancelKeyPress(object sender, ConsoleCancelEventArgs e)
        {
            DoQuit();
        }
        public static void DoQuit()
        {
            WriteCommandResponse("Shutting down...");
            if (null != oProxy) oProxy.Dispose();
            Fiddler.FiddlerApplication.Shutdown();
            Thread.Sleep(500);
        }
        public static void WriteCommandResponse(string s)
        {
            ConsoleColor oldColor = Console.ForegroundColor;
            Console.ForegroundColor = ConsoleColor.Yellow;
            Console.WriteLine(s);
            Console.ForegroundColor = oldColor;
        }
    }
}

處理類：

using Fiddler;
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Text.RegularExpressions;
using System.Threading;
using System.Threading.Tasks;
using System.Configuration;
using System.Xml;


namespace FiddlerDemo
{
    public class HtmlConfig
    {
        public string sWeb { get; set; }
        public string sRegex { get; set; }
    }
    public class FiddlerHelp
    {
        public static Proxy oProxy;
        public static List<HtmlConfig> dicHtml = new List<HtmlConfig> { };
     
        public void StartSession()
        {
            LoadHtmlConfig();
            //設置別名
            Fiddler.FiddlerApplication.SetAppDisplayName("FiddlerCoreDemoApp");

            List<Fiddler.Session> oAllSessions = new List<Fiddler.Session>();

            Fiddler.FiddlerApplication.OnNotification += delegate(object sender, NotificationEventArgs oNEA) { Console.WriteLine("** NotifyUser: " + oNEA.NotifyString); };

            Fiddler.FiddlerApplication.BeforeRequest += delegate(Fiddler.Session oS)
            {
                // 爲了使反應篡改，必須使用緩衝模式
                // 被啓用。這容許FiddlerCore以容許修改
                // 在BeforeResponse處理程序中的反應，而不是流
                // 響應給客戶機做爲響應進來。
                oS.bBufferResponse = true;
                Monitor.Enter(oAllSessions);
                oAllSessions.Add(oS);
                Monitor.Exit(oAllSessions);
            };
           
            // 下面的事件，您能夠檢查由Fiddler閱讀每一響應緩衝區。  
            // 請注意，這不是爲絕大多數應用很是有用，由於原始緩衝區幾乎是無用的;它沒有解壓，它包括標題和正文字節數等。
            // 本次僅適用於極少數的應用程序這就須要一個原始的，未經處理的字節流獲取有用
            //Fiddler.FiddlerApplication.OnReadResponseBuffer += new EventHandler<RawReadEventArgs>(FiddlerApplication_OnReadResponseBuffer);
            Fiddler.FiddlerApplication.BeforeResponse += delegate(Fiddler.Session oS)
            {
                //HTTP響應，並隨後修改任何HTTP響應，以取代
                oS.utilDecodeResponse();
                foreach (var item in dicHtml)
                {
                    if (oS.fullUrl.Contains(item.sWeb))
　　　　　　　　　　　　{
                        Console.WriteLine("{0}:HTTP {1} for {2}", oS.id, oS.responseCode, oS.fullUrl);
                        string sHtmlBody = oS.GetResponseBodyAsString();
                        if(!string.IsNullOrEmpty(sHtmlBody))
                        {
                            Console.Write("獲取的內容爲："+MatchRegex(sHtmlBody,item.sRegex) + "\n");
                        }
                    }
                }
                Monitor.Enter(oAllSessions);
                oAllSessions.Add(oS);
                Monitor.Exit(oAllSessions);
                // 內容：{3} , oS.GetResponseBodyEncoding().GetString(oS.responseBodyBytes)
                //Console.WriteLine("{0}:HTTP {1} for {2}", oS.id, oS.responseCode, oS.fullUrl);
            };

            Fiddler.FiddlerApplication.AfterSessionComplete += delegate(Fiddler.Session oS)
            {
                Console.Title = ("Session list contains: " + oAllSessions.Count.ToString() + " sessions");
 　　　　　　　　　//數量大於1000條時進行清空
                if (oAllSessions.Count > 1000)
                {
                    Monitor.Enter(oAllSessions);
                    oAllSessions.Clear();
                    Monitor.Exit(oAllSessions);
                }
            };
            Console.CancelKeyPress += new ConsoleCancelEventHandler(Console_CancelKeyPress);

            FiddlerApplication.Prefs.SetBoolPref("fiddler.network.streaming.abortifclientaborts", true);
            //啓動方式
            //FiddlerCoreStartupFlags oFCSF = FiddlerCoreStartupFlags.Default;
            Fiddler.CONFIG.IgnoreServerCertErrors = false;
            int iPort = 8877;
            //Fiddler.FiddlerApplication.Startup(iPort, oFCSF);
            Fiddler.FiddlerApplication.Startup(iPort, true, false, true);
            bool bDone = false;
            #region 各類操做
            do
            {
                Console.WriteLine("\nEnter a command [C=Clear; L=List; G=Collect Garbage; W=write SAZ; R=read SAZ;\n\tS=Toggle Forgetful Streaming; T=Trust Root Certificate; Q=Quit]:>");
                ConsoleKeyInfo cki = Console.ReadKey();
                Console.WriteLine();
                switch (Char.ToLower(cki.KeyChar))
                {
                    case 'c':
                        Monitor.Enter(oAllSessions);
                        oAllSessions.Clear();
                        Monitor.Exit(oAllSessions);
                        WriteCommandResponse("Clear...");
                        FiddlerApplication.Log.LogString("Cleared session list.");
                        break;

                    case 'd':
                        FiddlerApplication.Log.LogString("FiddlerApplication::Shutdown.");
                        FiddlerApplication.Shutdown();
                        break;

                    //case 'l':
                    //    WriteSessionList(oAllSessions);
                    //    break;

                    case 'g':
                        Console.WriteLine("Working Set:\t" + Environment.WorkingSet.ToString("n0"));
                        Console.WriteLine("Begin GC...");
                        GC.Collect();
                        Console.WriteLine("GC Done.\nWorking Set:\t" + Environment.WorkingSet.ToString("n0"));
                        break;

                    case 'q':
                        bDone = true;
                        DoQuit();
                        break;

                    case 'r':

                        WriteCommandResponse("This demo was compiled without SAZ_SUPPORT defined");

                        break;

                    case 'w':

                        WriteCommandResponse("This demo was compiled without SAZ_SUPPORT defined");
                        break;

                    case 't':
                        try
                        {
                            WriteCommandResponse("Result: " + Fiddler.CertMaker.trustRootCert().ToString());
                        }
                        catch (Exception eX)
                        {
                            WriteCommandResponse("Failed: " + eX.ToString());
                        }
                        break;

                    // Forgetful streaming
                    case 's':
                        bool bForgetful = !FiddlerApplication.Prefs.GetBoolPref("fiddler.network.streaming.ForgetStreamedData", false);
                        FiddlerApplication.Prefs.SetBoolPref("fiddler.network.streaming.ForgetStreamedData", bForgetful);
                        Console.WriteLine(bForgetful ? "FiddlerCore will immediately dump streaming response data." : "FiddlerCore will keep a copy of streamed response data.");
                        break;

                }
            } while (!bDone);
            #endregion
        }

        /// <summary>
        /// 經過讀取xml內的配置來獲取監聽的網站和獲取數據的正則表達式
        /// </summary>
        private void LoadHtmlConfig()
        {
            try
            {
                XmlDocument xmlDoc = new XmlDocument();
                 string sPath = string.Empty;
                if (File.Exists(@"..\..\WatchHtml.xml"))
                {
                    //調試目錄
                    sPath = @"..\..\WatchHtml.xml";
                }
                else
                {
                    //編譯目錄下
                    sPath = @"WatchHtml.xml";
                }
                XmlNode xn = xmlDoc.SelectSingleNode("Root");
                XmlNodeList xnl = xn.ChildNodes;
                foreach (XmlNode item in xnl)
                {
                    XmlElement xe = (XmlElement)item;
                    HtmlConfig htmlConfig = new HtmlConfig
                    {
                        sWeb = xe.GetAttribute("Web").ToString(),
                        sRegex = xe.GetAttribute("Regex").ToString()
                    };
                    dicHtml.Add(htmlConfig);
                }
            }
            catch (Exception)
            {

                throw;
            }
        }

        public void Console_CancelKeyPress(object sender, ConsoleCancelEventArgs e)
        {
            DoQuit();
        }
        /// <summary>
        /// 退出
        /// </summary>
        public void DoQuit()
        {
            if (null != oProxy) oProxy.Dispose();
            Fiddler.FiddlerApplication.Shutdown();
            Thread.Sleep(500);
        }
        public static void WriteCommandResponse(string s)
        {
            ConsoleColor oldColor = Console.ForegroundColor;
            Console.ForegroundColor = ConsoleColor.Yellow;
            Console.WriteLine(s);
            Console.ForegroundColor = oldColor;
        }
       /// <summary>
       /// 
       /// </summary>
       /// <param name="sHtml">得到的Html頁面</param>
       /// <param name="sRegex">正則表達式</param>
       /// <returns></returns>
        public static string MatchRegex(string sHtml,string sRegex)
        {
            string sResult = string.Empty;
            try
            {
                if (string.IsNullOrEmpty(sHtml))
                    return null;
                var result = Regex.Match(sHtml.Replace('\r', ' ').Replace('\n', ' ').Trim(), sRegex, RegexOptions.IgnoreCase | RegexOptions.Multiline);
                if (result.Success)
                {
                    sResult = result.Value;
                }
                return sResult;
            }
            catch (Exception)
            {
                return null;
            }
        }

    }
}

XML文件內容

實現效果

 

 

 

 

 

 

---恢復內容結束---