What is the fastest way for checking a big proxy list on a specific web site?

问题

I have a big list of proxy servers (txt file , Format = ip:port in each line) and wrote the code below for checking them:

    public static void MyChecker()
    {
        string[] lines = File.ReadAllLines(txtProxyListPath.Text);
        List<string> list_lines = new List<string>(lines);
        List<string> list_lines_RemovedDup = new List<string>();
        HashSet<string> HS = new HashSet<string>();
        int Duplicate_Count = 0;
        int badProxy = 0;
        int CheckedCount = 0;

        foreach (string line in list_lines)
        {
            string[] line_char = line.Split(':');
            string ip = line_char[0];
            string port = line_char[1];
            if (CanPing(ip))
            {
                if (SoketConnect(ip, port))
                {
                    if (CheckProxy(ip, port))
                    {
                        string ipAndport = ip + ":" + port;
                        if (HS.Add(ipAndport))
                        {
                            list_lines_RemovedDup.Add(ipAndport);
                            CheckedCount++;
                        }
                        else
                        {
                            Duplicate_Count++;
                            CheckedCount++;
                        }
                    }
                    else
                    {
                        badProxy++;
                        CheckedCount++;
                    }
                }
                else
                {
                    badProxy++;
                    CheckedCount++;
                }
            }
            else
            {
                badProxy++;
                CheckedCount++;
            }
    }

    public static bool CanPing(string ip)
    {
        Ping ping = new Ping();

        try
        {
            PingReply reply = ping.Send(ip, 2000);
            if (reply == null)
                return false;

            return (reply.Status == IPStatus.Success);
        }
        catch (PingException Ex)
        {
            return false;
        }
    }

    public static bool SoketConnect(string ip, string port)
    {
        var is_success = false;
        try
        {
            var connsock = new Socket(AddressFamily.InterNetwork, SocketType.Stream, ProtocolType.Tcp);
            connsock.SetSocketOption(SocketOptionLevel.Socket, SocketOptionName.SendTimeout, 200);
            System.Threading.Thread.Sleep(500);
            var hip = IPAddress.Parse(ip);
            var ipep = new IPEndPoint(hip, int.Parse(port));
            connsock.Connect(ipep);
            if (connsock.Connected)
            {
                is_success = true;
            }
            connsock.Close();
        }
        catch (Exception)
        {
            is_success = false;
        }
        return is_success;
    }

    public static bool CheckProxy(string ip, string port)
    {
        try
        {
            WebClient WC = new WebClient();
            WC.Proxy = new WebProxy(ip, int.Parse(port));
            WC.DownloadString("http://SpecificWebSite.com");
            return true;
        }
        catch (Exception)
        {
            return false;
        }
    }

But I think I should rewrite these codes because they are very slow.
I have bad delays in these lines :
WC.DownloadString("http://SpecificWebSite.com");
and
PingReply reply = ping.Send(ip, 2000);
and this is not good for a big list.
Did I write these codes in the right direction or should i change them(which parts)?
how can i optimze them?

thanks in advance

回答1:

There are quite a few things you can improve.

• Don't sleep the thread for half a second.
• Drop the ping check (since the proxy might be behind a firewall and not responding to pings but still working)
• Replace DownloadString with a HttpWebRequest getting the HEAD only.
• Set the timeout of your HttpWebRequest to something lower than default (no need to wait that long. If a proxy doesn't respond within 10-20secs then you probably don't want to use it).
• Split your big list into smaller ones and process them at the same time.

These alone should speed up your process by quite a bit.

As requested, here's an example of how to use HttpWebRequests

HttpWebRequest request = (HttpWebRequest)WebRequest.Create(url);
request.Proxy = null;   // set proxy here
request.Timeout = 10000; 
request.Method = "HEAD";

using (HttpWebResponse response = (HttpWebResponse)request.GetResponse())
{
    Console.WriteLine(response.StatusCode);
}

回答2:

I might do something like this:

public static bool TestProxy(string ipAddress, int port, out string errorMsg, out double connectionSeconds) {
  Stopwatch stopWatch = new Stopwatch();
  errorMsg = "";
  connectionSeconds = -1;

  try {
    stopWatch.Start();
    var client = new RestClient("https://webapi.theproxisright.com/");
    client.Proxy = new WebProxy(ipAddress, port);

    var request = new RestRequest("api/ip", Method.GET);
    request.Timeout = 10000;
    request.RequestFormat = DataFormat.Json;

    var response = client.Execute(request);
    if (response.ErrorException != null) {
      throw response.ErrorException;
    }
    return (response.Content == ipAddress);
  } catch (Exception ex) {
    errorMsg = ex.Message;
    return false;
  } finally {
    stopWatch.Stop();
    connectionSeconds = stopWatch.Elapsed.TotalSeconds;
  }
}

Using a WhatIsMyIP-like REST service (I use one from https://TheProxIsRight.com).

Then As suggested above, I might try parallelize it with something like:

  Task.Factory.StartNew(() => {
    try {
      string errorMsg;
      double connectionTime;
      var success = TestProxy("1.2.3.4",3128, out errorMsg, out connectionTime);

      //Log Result
    } catch (Exception ex) {
      //Log Error
    }
  });

Note, one can also use the REST API on the above site to query for working proxies: https://theproxisright.com/#apidemo

(Disclosure, I worked on the above site)

来源：https://stackoverflow.com/questions/12249702/what-is-the-fastest-way-for-checking-a-big-proxy-list-on-a-specific-web-site