使用HttpWebRequest模拟登陆阿里巴巴(alibaba、httpwebrequest、login)

前言

　　其实老喜欢取经，偶尔也得分享下。关于阿里巴巴国际站的登陆，过程有点复杂但是算不上难。一不小心少个东西倒也挺麻烦的。

　　主要是看下请求类HttpClient基本请求封装使用,AliClient模拟浏览器的操作与数据封装

　　这里只是简单说一下主要的类和注意点，主要步骤与注意点都写在代码注释里了。项目源码下载地址:http://git.oschina.net/GspringG/AliLogin

正文

1. 主要类/方法

• HttpClient请求模拟的基础类，也就那么个过程http header设置一下，然后模拟就行了.需要注意的地方在代码注释里

  using System;
  using System.Drawing;
  using System.IO;
  using System.Net;
  using System.Net.Security;
  using System.Security.Cryptography.X509Certificates;
  using System.Text;
  
  namespace Main.Http
  {
      public class HttpClient
      {
          const string Accept = "text/html, application/xhtml+xml, */*";
          public const string UserAgent = "Mozilla/5.0 (Windows NT 6.1; WOW64; Trident/7.0; rv:11.0) like Gecko";
          const int Timeout = 15000;//超时时间，但前提是请求允许超时才行。
          const string AcceptLanguage = "Accept-Language:zh-CN";
          const string AcceptEncoding = "Accept-Encoding:gzip, deflate";
          const string Contenttype = "application/x-www-form-urlencoded";
          /// <summary>
          /// 静态构造方法
          /// </summary>
          static HttpClient()
          {
              //伪造证书,验证服务器证书回调自动验证
              ServicePointManager.ServerCertificateValidationCallback = CheckValidationResult;
              //客户端系统 win7或者winxp上可能会出现 could not create ssl/tls secure channel的问题导致加载ali登陆验证码报错
              //这里必须设置
              ServicePointManager.Expect100Continue = false;//默认是true，要手动设为false 
              ServicePointManager.SecurityProtocol=SecurityProtocolType.Ssl3;
  
              ServicePointManager.DefaultConnectionLimit = 1000;
          }
  
          public static string Get(string url, CookieContainer cookie, int retryCount = 3)
          {
              return Request(url, cookie, "GET", null, retryCount);
          }
          public static string Post(string url, CookieContainer cookie, string postData, int retryCount = 3)
          {
              return Request(url, cookie, "POST", postData, retryCount);
          }
  
          public static string Post(string url, CookieContainer cookie, string postData, Action<HttpWebRequest> beginRequest, int reTry = 0)
          {
              return Request(url, cookie, "POST", postData, reTry, beginRequest);
          }
  
          private static string Request(string url, CookieContainer cookie, string method, string postData, int retryCount, Action<HttpWebRequest> beginRequest = null)
          {
              string html = string.Empty;
              for (var i = 0; i <= retryCount; i++)
              {
                  try
                  {
                      html = Request(url, cookie, method, postData, beginRequest);
                      return html;
                  }
                  catch (Exception e)
                  {
                      if (i == retryCount)
                      {
                          throw new Exception(e.ToString());
                      }
                  }
              }
              return html;
          }
  
          public static bool CheckValidationResult(object sender, X509Certificate certificate, X509Chain chain, SslPolicyErrors errors)
          {   // 伪造证书,总是接受  
              return true;
          }
  
          private static string Request(string url, CookieContainer cookie, string method, string postData, Action<HttpWebRequest> beginRequestHandle = null)
          {
              var request = (HttpWebRequest)WebRequest.Create(url);
              request.Method = method;
              request.CookieContainer = cookie;
              request.AllowAutoRedirect = true;
              request.ContentType = Contenttype;
              request.Accept = Accept;
              request.Timeout = Timeout;
              request.UserAgent = UserAgent;
              request.Headers.Add(AcceptLanguage);
              request.Headers.Add(AcceptEncoding);
  
              request.AutomaticDecompression = DecompressionMethods.GZip;
  
              if (beginRequestHandle != null)
                  beginRequestHandle(request);
  
              if (!string.IsNullOrEmpty(postData))
              {
                  byte[] byteRequest = Encoding.UTF8.GetBytes(postData);
                  request.ContentLength = byteRequest.Length;
                  var stream = request.GetRequestStream();
                  stream.Write(byteRequest, 0, byteRequest.Length);
                  stream.Close();
              }
              var httpWebResponse = (HttpWebResponse)request.GetResponse();
  
              var responseStream = httpWebResponse.GetResponseStream();
              if (responseStream == null) return string.Empty;
              if (responseStream.CanTimeout)
              {
                  responseStream.ReadTimeout = 15000;
              }
              var encoding = Encoding.GetEncoding(httpWebResponse.CharacterSet ?? "UTF8");
  
              var streamReader = new StreamReader(responseStream, encoding);
  
              string html = streamReader.ReadToEnd();
              streamReader.Close();
              responseStream.Close();
              request.Abort();
              httpWebResponse.Close();
              return html;
          }
  
          public static byte[] DownloadFile(string url, CookieContainer cookie)
          {
              var webClient = new CookieWebClient { Cookie = cookie };
              byte[] data = webClient.DownloadData(url);
              return data;
          }
          public static Image DownloadImage(string url, CookieContainer cookie)
          {
              byte[] data = DownloadFile(url, cookie);
              var ms = new MemoryStream(data);
              var image = Image.FromStream(ms);
              ms.Close();
              return image;
          }
      }
  }

• AliClient:相当于一个浏览器，存储着cookie、session、_csrf_token_、userName等信息。客户端发出的模拟请求都通过AliClient发出保证全部在一个会话中。(想想一个浏览器需要些什么，浏览器是怎么做的。就感觉很合理了)

  using System;
  using System.Collections.Specialized;
  using System.Net;
  using System.Web;
  using Main.Http;
  
  namespace Main.Ali
  {
      public class AliClient
      {
          const string _csrf_token_ = "_csrf_token_";
          public CookieContainer Cookie { get; set; }
          public string SessionId { get; set; }
  
          public string CsrfToken { get; set; }
          public AliLoginUser AliLoginUser { get; set; }
          public string DmtrackPageid { get; set; }
          public string UserName
          {
              get
              {
                  if (AliLoginUser == null) return null;
                  if (AliLoginUser.person_data == null) return null;
                  return AliLoginUser.person_data.login_id;
              }
          }
          public AliClient()
          {
              Cookie = new CookieContainer();
          }
          public string Get(string url)
          {
              return HttpClient.Get(AddCsrfTokenToUrl(url), this.Cookie, 3);
          }
          public string Post(string url, string postData)
          {
              return HttpClient.Post(AddCsrfTokenToUrl(url), this.Cookie, AddCsrfTokenToPostData(postData), 3);
          }
          //post请求时不用手动加了，默认都加上
          private string AddCsrfTokenToPostData(string postData)
          {
              NameValueCollection queryString = HttpUtility.ParseQueryString(postData);
              if (queryString[_csrf_token_] == null)
              {
                  queryString.Add("_csrf_token_", this.CsrfToken);
              }
              return queryString.ToString();
          }
          //Get请求时不用手动加了，默认都加上
          private string AddCsrfTokenToUrl(string url)
          {
  
              if (string.IsNullOrWhiteSpace(this.CsrfToken)) return url;
              UriBuilder ub = new UriBuilder(url);
              NameValueCollection queryString = HttpUtility.ParseQueryString(ub.Query);
              if (queryString[_csrf_token_] == null)
              {
                  queryString.Add("_csrf_token_", this.CsrfToken);
              }
              ub.Query = queryString.ToString();
              string newurl = ub.Uri.ToString();
              return newurl;
          }
  
      }
  }

• AliPassporter：阿里巴巴登陆核心类(具体个注意细节点都在代码内部已注释)

　　　　登陆过程为:从上到下依次

　　　　　　1.PrepareLogin:

        /// <summary>
        /// 登陆前先请求一下页面获取sessionId,DmtrackPageid,初始化cookie的值,像正常访问浏览器一样
        /// </summary>
        /// <param name="aliClient"> AliClient当需要多账号登陆时，每个账号的cookie 各种验证id等独有的东西放到各自的Client中，相当于每个独立的浏览器</param>
        /// <param name="isSpec">针对某种特殊情况取值方法不一样，默认false</param>
        public static void PrepareLogin(AliClient aliClient, bool isSpec = false);

　　　　　　2.DoLoginStep1:

       /// <summary>
        ///     登录阿里巴巴
        /// </summary>
        /// <param name="aliClient"> AliClient当需要多账号登陆时，每个账号的cookie 各种验证id等独有的东西放到各自的Client中</param>
        /// <param name="account">帐户名</param>
        /// <param name="password">密码</param>
        /// <param name="checkCode">验证码</param>
        /// <returns>登陆结果(true/false)</returns>
        public static bool DoLoginStep1(AliClient aliClient, string account, string password, string checkCode);

　　　　　　3.DoLoginStep2:

        /// <summary>
        ///     登录处理第二步
        /// </summary>
        /// <param name="aliClient"></param>
        /// <param name="userId">账户</param>
        /// <param name="password">密码</param>
        /// <param name="dmtrackPageid">令牌1</param>
        /// <param name="st">令牌三</param>
        /// <returns>登陆结果</returns>
        private static bool DoLoginStep2(AliClient aliClient, string userId, string password, string dmtrackPageid, string st);

　　　　　　4.DoCheckCode:

        /// <summary>
        /// 获取checkcode，图片验证码，其实有时候是不需要的。可以先判断一下，不行再获取验证码
        /// </summary>
        /// <param name="aliClient">aliClient</param>
        /// <returns>验证码图片，如果是web应用，直接把checkCodeUrl(img src=checkCodeUrl)写进去就行了</returns>
        public static Image DoCheckCode(AliClient aliClient);

　　　　　　5.登陆过程中还会调用GetToken、GetSt、GetCsrfToken等方法提供登录所需要的get/post参数

　　　　　　6.前端调用流程:登录过程关键代码(详情参见详细代码)

           //1.初始化_aliClient，相当于打开一个浏览器，并设置一个空的cookie
           AliClient  _aliClient=new AliClient {Cookie = new CookieContainer()};
           //2.无需验证码登陆时这里替换为PrepareLogin(_aliClient);
           RefreshCheckCode();
           //3.登陆系统，异步方法防止页面假死
           var data=await AliPassporter.DoLoginAsync(_aliClient, UserName, Password, CheckCode);
           if (data)
            {
                //登陆成功,登陆后其他数据就随便抓了,这里要注意，需要手机验证码的，要在网页中先把手机验证码输入了
                //我的测试账号默认写在里面，别乱玩就行了
                //进入管理个人信息页面
                var html = _aliClient.Get("http://accounts.alibaba.com/user/organization/manage_person_profile.htm");
                MessageBox.Show("登陆成功!");
                //_aliClient.Post("");
            }
            else
            {
                MessageBox.Show("登陆失败!");
            }            

 总结

　　以上就是阿里巴巴国际站登陆的全部过程，总体大同小异，很久之前写的了(建议先把注释看一遍再用)。现在拿出来分享一下，顺便复习一下HttpWebRequest的相关使用。