Java 登录百度并获取 Cookie

勤为径苦作舟

已于 2023-09-27 11:49:24 修改

阅读量3.4k

点赞数 1

文章标签：爬虫

于 2018-04-13 10:22:21 首次发布

本文链接：https://blog.csdn.net/duanluan/article/details/79924570

版权

package top.duanluan;

import com.gargoylesoftware.htmlunit.BrowserVersion;
import com.gargoylesoftware.htmlunit.WebClient;
import com.gargoylesoftware.htmlunit.html.DomElement;
import com.gargoylesoftware.htmlunit.html.HtmlAnchor;
import com.gargoylesoftware.htmlunit.html.HtmlPage;
import com.gargoylesoftware.htmlunit.util.Cookie;

import java.util.List;
import java.util.Set;

public class Test {

    public static void main(String[] args) throws Exception {
        String username = "";
        String password = "";

        WebClient webClient = new WebClient(BrowserVersion.CHROME);
        // 等待后台 JS 执行
        webClient.waitForBackgroundJavaScript(10000);
        HtmlPage page = webClient.getPage("http://developer.baidu.com/");

        List<HtmlAnchor> loginAnchors = page.getByXPath("//*[@id='navbar']/ul[2]/li[2]/a");
        page = loginAnchors.get(0).click();

        // 获取用户名和密码 Input 并输入
        DomElement usernameInput = null;
        DomElement passwordInput = null;
        for (int i = 0; i < 20; i++) {
            usernameInput = page.getElementById("TANGRAM__PSP_10__userName");
            if (usernameInput != null) {
                break;
            }
            synchronized (page) {
                page.wait(500);
            }
        }
        passwordInput = page.getElementById("TANGRAM__PSP_10__password");
        usernameInput.setAttribute("value", username);
        passwordInput.setAttribute("value", password);

        // 获取并点击登录按钮
        DomElement loginBtn = page.getElementById("TANGRAM__PSP_10__submit");
        loginBtn.click();

        // 获取 Cookie
        Set<Cookie> cookies = webClient.getCookieManager().getCookies();
        for (Cookie cookie : cookies) {
            System.out.println(cookie.getName() + ": " + cookie.getValue());
        }
    }
}