C#自动登录网页浏览页面 抓取数据

需求：客户的数据同时存在在另外一个不可控的系统中，需要和当前系统同步。

思路：自动登录另外一个系统，然后抓取数据，同步到本系统中。

技术点：模拟用户登录；保存登录状态；抓取数据

 

程序非常简单：

     /// <summary>

        /// visit the target url

        /// </summary>

        /// <param name="targetURL"></param>

        /// <param name="cc">this is for keeping cookies and sessions</param>

        /// <param name="param">this is the data need post inside form</param>

        /// <returns>html page</returns>

        public static string PostAndGetHTML(string targetURL,CookieContainer cc, Hashtable param)

        {

            //prepare the submit data

            string formData = "";

            foreach (DictionaryEntry de in param)

            {

                formData += de.Key.ToString() + "=" + de.Value.ToString()+"&";

            }

            if(formData.Length>0)

               formData = formData.Substring(0, formData.Length - 1); //remove last '&'

 

            ASCIIEncoding encoding = new ASCIIEncoding();

            byte[] data = encoding.GetBytes(formData);

 

            HttpWebRequest request = (HttpWebRequest)WebRequest.Create(targetURL);

            request.Method = "POST";    //post

            request.ContentType = "application/x-www-form-urlencoded";

            request.ContentLength = data.Length;

            request.UserAgent = "Mozilla/4.0 (compatible; MSIE 7.0; Windows NT 5.1; SV1; .NET CLR 2.0.1124)";

            

            Stream newStream = request.GetRequestStream();

            newStream.Write(data, 0, data.Length);

 

            newStream.Close();

 

            request.CookieContainer = cc;

            HttpWebResponse response = (HttpWebResponse)request.GetResponse();

            cc.Add(response.Cookies);

            Stream stream = response.GetResponseStream();

            string result = new StreamReader(stream, System.Text.Encoding.Default).ReadToEnd();

            return result;

        }

这一个是调用的例子：先登录，在查询。 实际中这个逻辑可能有很多步骤

        private void button2_Click(object sender, EventArgs e)

        {

            CookieContainer cc = new CookieContainer();//this is for keep the Session and Cookie

            Hashtable param = new Hashtable();//this is for keep post data.

 

            string urlLogin = "http://demo.server//login.asp";

             //do find the elementId that needed. check the source of login page can get this information

            param.Add("User", "xxx");

            param.Add("Password", "xxxx");

            string result = PostAndGetHTML(urlLogin, cc, param);

            //check result, whether login success

          

            //if login success, goto the target url, and input some value.

            string url2 = " http://demo.server/query.asp?id=1";// need change. special logic

            param.Clear();

            //param.Add("SearchAreaId","JobId")

            result = PostAndGetHTML(url2, cc, new Hashtable());

            //ConvertToDT the html or do something others

 

 

        }

这是一个简单的抓取网页数据的函数（针对Table内的，直接转化成DataTable

）
        private DataTable ConvertToDT(DataTable dt, string tableHTML)

        {

 

            int lastTD = tableHTML.ToLower().LastIndexOf("</td>");

            int firstRow = tableHTML.ToLower().IndexOf("<tr") + 3;//after ""<tr

            int index = tableHTML.ToLower().IndexOf("<tr", firstRow) + 3;//after ""<tr

            while (index < lastTD)

            {

                DataRow dr = dt.NewRow();

                for (int i = 0; i < dt.Columns.Count; i++)

                {

                    string value = "";

                    int startTD = tableHTML.ToLower().IndexOf("<td", index) + 3;//after "<td"

                    int endTD = tableHTML.ToLower().IndexOf("</td>", startTD);

                    if (endTD < 0)

                        break;

                    string tdStr = tableHTML.Substring(startTD, endTD - startTD);

                   

                    //remove <> and others

                    tdStr = tdStr.Replace("&nbsp;", "").Replace("\t", "").Replace("\r", "");

                    string[] v = tdStr.Split('<', '>');

                    for (int j = 0; j < v.Length; j++)

                    {

                        j++;

                        if (v[j].Trim() != "")

                        {

                            value = v[j].Trim();

                            break;

                        } 

                    }

                    //

                    dr[i] = value;

                    index = endTD;

                }

                dt.Rows.Add(dr);

 

            }

            return dt;

        }

注：对于有验证码登录系统的无效。（如果该系统的验证码放到cookie中存储的例外，这个容易破解）