背景
在1688官网里面有很多信息是需要登录才能看得到的,比如商家的联系电话等等。那么我们在抓取它的网页的时候,肯定是需要维持登录状态才能得到对应的内容。这里面就会涉及到自动登录的问题。
登录地址
https://login.1688.com/member/signin.htm
自动登录方法
1、找到对应的元素,账号、密码框。
2、把账号、密码值带进去。
List<string> logininfolist = new List<string>();string file = "1688Account.json";if (!File.Exists(file)){throw new ArgumentException("1688Account not found");}string data = File.ReadAllText(file, Encoding.UTF8);var account1688list = JsonConvert.DeserializeObject<List<Account1688Item>>(data);logininfolist.AddRange(account1688list.Select(o => o.AccountPassword).ToList());Random rdinfo = new Random();int indexinfo = rdinfo.Next(logininfolist.Count);var modelinfo = logininfolist[indexinfo];driver.FindElement(By.Id("fm-login-id")).SendKeys(modelinfo.Split('+')[0]);driver.FindElement(By.Id("fm-login-password")).SendKeys(modelinfo.Split('+')[1]);
3、模拟点击提交按钮操作。
// Thread.Sleep(1000 * 30); //30时间操作driver.FindElement(By.ClassName("password-login")).Click();Thread.Sleep(1000 * 30);//30时间操作
4、进入控制台,验证是否登陆成功。并记录Cookies,下次再来的时候,直接使用现成的Cookies,一般有效期可以维持一天多左右。
driver.Navigate().GoToUrl("https://work.1688.com/?tracelog=login_target_is_blank_1688");Thread.Sleep(1000 * 5);if (driver.Url.Contains("login.1688.com")){Console.WriteLine("登录失败");CookieHelp.DeleteCookies();Console.WriteLine("2");throw new Exception("重新登录");}driver.Navigate().Refresh();CookieHelp.WriteCookies(driver.Manage().Cookies.AllCookies);
5、判断是否有现成的登录cookies。
driver.Navigate().GoToUrl("https://www.1688.com/");driver.Manage().Cookies.DeleteAllCookies();var listCookie = CookieHelp.GetCookie();if (listCookie != null){logintry = 0;Console.WriteLine("有现成cookies" + DateTime.UtcNow);foreach (var item in listCookie){driver.Manage().Cookies.AddCookie(new Cookie(item.Name, item.Value, item.Domain, item.Path, item.Expiry));}Thread.Sleep(2000);driver.Navigate().GoToUrl("https://work.1688.com/?tracelog=login_target_is_blank_1688");Thread.Sleep(1000 * 2);
完整代码
/// <summary>/// 登录 todo/// </summary>/// <param name="_reptilesImageSearchService"></param>/// <param name="options"></param>/// <param name="driver"></param>public void Implement(IReptilesImageSearchService _reptilesImageSearchService, IWebDriver driver){driver.Navigate().GoToUrl("https://www.1688.com/");driver.Manage().Cookies.DeleteAllCookies();var listCookie = CookieHelp.GetCookie();if (listCookie != null){logintry = 0;Console.WriteLine("有现成cookies" + DateTime.UtcNow);foreach (var item in listCookie){driver.Manage().Cookies.AddCookie(new Cookie(item.Name, item.Value, item.Domain, item.Path, item.Expiry));}Thread.Sleep(2000);driver.Navigate().GoToUrl("https://work.1688.com/?tracelog=login_target_is_blank_1688");Thread.Sleep(1000 * 2);if (driver.Url.Contains("login.1688.com")){Console.WriteLine("cookies过期了");CookieHelp.DeleteCookies();Console.WriteLine("1");throw new Exception("重新登录");}}else{if (logintry > 4){Console.WriteLine("登陆次数超出:" + logintry);throw new Exception("登陆次数超出,退出");}logintry++;Console.WriteLine("无现成cookies" + DateTime.UtcNow);driver.Navigate().GoToUrl("https://login.1688.com/member/signin.htm");#region 登录动作driver.SwitchTo().Frame(0);IJavaScriptExecutor js = (IJavaScriptExecutor)driver;// string returnjs = (string)js.ExecuteScript("Object.defineProperties(navigator, {webdriver:{get:()=>undefined}});");string jsfile = File.ReadAllText("stealth.min.js", Encoding.UTF8);string returnjs = (string)js.ExecuteScript(jsfile);List<string> logininfolist = new List<string>();string file = "1688Account.json";if (!File.Exists(file)){throw new ArgumentException("1688Account not found");}string data = File.ReadAllText(file, Encoding.UTF8);var account1688list = JsonConvert.DeserializeObject<List<Account1688Item>>(data);logininfolist.AddRange(account1688list.Select(o => o.AccountPassword).ToList());Random rdinfo = new Random();int indexinfo = rdinfo.Next(logininfolist.Count);var modelinfo = logininfolist[indexinfo];driver.FindElement(By.Id("fm-login-id")).SendKeys(modelinfo.Split('+')[0]);driver.FindElement(By.Id("fm-login-password")).SendKeys(modelinfo.Split('+')[1]);// Thread.Sleep(1000 * 30); //30时间操作driver.FindElement(By.ClassName("password-login")).Click();Thread.Sleep(1000 * 30);//30时间操作#endregion 登录动作driver.Navigate().GoToUrl("https://work.1688.com/?tracelog=login_target_is_blank_1688");Thread.Sleep(1000 * 5);if (driver.Url.Contains("login.1688.com")){Console.WriteLine("登录失败");CookieHelp.DeleteCookies();Console.WriteLine("2");throw new Exception("重新登录");}driver.Navigate().Refresh();CookieHelp.WriteCookies(driver.Manage().Cookies.AllCookies);}Thread.Sleep(1000);}
注意事项
1、登录的时候,有时候会出现滑块验证码,这时候一般是使用其他账号重试或者是在当前的机器手工登录一次,后面基本就会被信任。
2、使用这段代码本身就是模拟真实用户的行为,最大限度的减少验证码出现的几率。
IJavaScriptExecutor js = (IJavaScriptExecutor)driver;// string returnjs = (string)js.ExecuteScript("Object.defineProperties(navigator, {webdriver:{get:()=>undefined}});");string jsfile = File.ReadAllText("stealth.min.js", Encoding.UTF8);string returnjs = (string)js.ExecuteScript(jsfile);