無論使用任何語言做模擬登陸或者抓取訪問頁面,無外乎以下思路: 觀察這三個請求的詳細信息,不難看出第一個GET請求的地址可以由POST的響應得到,而第二個GET請求的地址又由第一個GET的響應得到。
HttpWebRequest request = null; HttpWebResponse response = null; string gethost = string.Empty; CookieContainer cc = new CookieContainer(); string Cookiesstr = string.Empty; try { //第一次POST請求 string postdata = "email=" + UserName.Replace("@", "%40") + "&password=" + PassWord + "&origURL=" + HostUrl + "&domain=renren.com";//模擬請求數(shù)據(jù),數(shù)據(jù)樣式可以用FireBug插件得到。人人網(wǎng)POST數(shù)據(jù)時,用戶名郵箱中的“@”變?yōu)椤?40”,所以我們也要作此變化 string LoginUrl="http://www.renren.com/PLogin.do"; request = (HttpWebRequest)WebRequest.Create(LoginUrl);//實例化web訪問類 request.Method = "POST";//數(shù)據(jù)提交方式為POST //模擬頭 request.ContentType = "application/x-www-form-urlencoded"; byte[] postdatabytes = Encoding.UTF8.GetBytes(postdata); request.ContentLength = postdatabytes.Length; //request.Referer = "http://www.renren.com/Login.do?rf=r&domain=renren.com&origURL=" + HostUrl; request.AllowAutoRedirect = false; request.CookieContainer = cc; request.KeepAlive = true; //提交請求 Stream stream; stream = request.GetRequestStream(); stream.Write(postdatabytes, 0, postdatabytes.Length); stream.Close(); //接收響應 response = (HttpWebResponse)request.GetResponse(); //保存返回cookie response.Cookies = request.CookieContainer.GetCookies(request.RequestUri); CookieCollection cook = response.Cookies; string strcrook = request.CookieContainer.GetCookieHeader(request.RequestUri); Cookiesstr = strcrook; //取第一次GET跳轉(zhuǎn)地址 StreamReader sr = new StreamReader(response.GetResponseStream(), Encoding.UTF8); string content = sr.ReadToEnd(); response.Close(); string[] substr = content.Split(new char[] { '"' }); gethost = substr[1]; } catch (Exception) { //第一次POST出錯; } 注釋寫的很詳細了,在這就不再分析,也許有人對request = (HttpWebRequest)WebRequest.Create(LoginUrl)有疑問,可以去google一下HttpWebRequest和WebRequest的區(qū)別,簡單來說WebRequest是一個抽象類,不能直接實例化,需要被繼承,而HttpWebRequest繼承自WebRequest。 再模擬第一個和第二個GET請求
try { request = (HttpWebRequest)WebRequest.Create(gethost); request.Method = "GET"; request.KeepAlive = true; request.Headers.Add("Cookie:" + Cookiesstr); request.CookieContainer = cc; request.AllowAutoRedirect = false; response = (HttpWebResponse)request.GetResponse(); //設(shè)置cookie Cookiesstr = request.CookieContainer.GetCookieHeader(request.RequestUri); //取再次跳轉(zhuǎn)鏈接 StreamReader sr = new StreamReader(response.GetResponseStream(), Encoding.UTF8); string ss = sr.ReadToEnd(); string[] substr = ss.Split(new char[] { '"' }); gethost = substr[1]; request.Abort(); sr.Close(); response.Close(); } catch (Exception) { //第一次GET出錯 } try { //第二次GET請求 request = (HttpWebRequest)WebRequest.Create(gethost); request.Method = "GET"; request.KeepAlive = true; request.Headers.Add("Cookie:" + Cookiesstr); request.CookieContainer = cc; request.AllowAutoRedirect = false; response = (HttpWebResponse)request.GetResponse(); //設(shè)置cookie Cookiesstr = request.CookieContainer.GetCookieHeader(request.RequestUri); request.Abort(); response.Close(); } catch (Exception) { //第二次GET出錯 } GET與POST請求大同小異,這里便不再累述。三次請求結(jié)束,保存好你的cookie string,每次請求的時候都賦給請求的頭部,你就處于登錄狀態(tài)了。 |