爬蟲任務三：模擬登陸博客園

httpclient模擬登陸

package com.zhaowu.renwu3;

import java.io.IOException;

import org.apache.commons.httpclient.Cookie;
import org.apache.commons.httpclient.HttpClient;
import org.apache.commons.httpclient.HttpException;
import org.apache.commons.httpclient.NameValuePair;
import org.apache.commons.httpclient.cookie.CookiePolicy;
import org.apache.commons.httpclient.methods.GetMethod;
import org.apache.commons.httpclient.methods.PostMethod;
import org.apache.commons.httpclient.params.HttpMethodParams;

public class BoKeYuanLogin {

    public static void main(String[] args) {
        String username = "******";
        String password = "******";
        String loginUrl = "https://passport.cnblogs.com/login.aspx";
        String targetUrl = "https://www.cnblogs.com/";

        HttpClientLogin(username, password, loginUrl, targetUrl);
    }

    private static void HttpClientLogin(String username, String password, String loginUrl, String targetUrl) {

        // 建立一個客戶端，相似於打開一個瀏覽器
        HttpClient httpClient = new HttpClient();
        httpClient.getParams().setParameter(HttpMethodParams.HTTP_URI_CHARSET, "utf-8");
        // 建立post方法實例,模擬登陸
        PostMethod postMethod = new PostMethod(loginUrl);
        // 設置登錄時要求的信息,即填入各個表單域的值，用戶名和密碼
        NameValuePair[] data = { new NameValuePair("username", username), 
　　　　　　　　　　　　　　　　　　　 new NameValuePair("password", password) };
        // 將表單的值放入到post方法中
        postMethod.setRequestBody(data);

        try {
            // 設置 HttpClient 接收 Cookie,用與瀏覽器同樣的策略
            httpClient.getParams().setCookiePolicy(CookiePolicy.BROWSER_COMPATIBILITY);
            // 得到響應狀態碼
            int postStatusCode = httpClient.executeMethod(postMethod);
            System.out.println("postStatusCode: " + postStatusCode);
            System.out.println("======================1==================");

            // 得到登錄後的 Cookie
            Cookie[] cookies = httpClient.getState().getCookies();
            System.out.println("cookies: " + cookies.toString());
            System.out.println("======================2==================");
            StringBuffer sb = new StringBuffer();
            for (Cookie cookie : cookies) {
                sb.append(cookie.toString() + ";");
                System.out.println(cookie.getName() + ":" + cookie.getValue());
                System.out.println("-------------------------");
            }
            /*System.out.println(sb.toString());
            System.out.println("======================5==================");*/

            GetMethod getMethod = new GetMethod(targetUrl);
            getMethod.setRequestHeader("Cookie", sb.toString());
            getMethod.setRequestHeader("Host", "passport.cnblogs.com");
            getMethod.setRequestHeader("Referer", "https://www.cnblogs.com/");
            getMethod.setRequestHeader("User-Agent",
                    "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:60.0) Gecko/20100101 Firefox/60.0");
            int getStatusCode = httpClient.executeMethod(getMethod);
            System.out.println("getStatusCode: " + getStatusCode);
            System.out.println("======================3==================");

            // 打印出返回數據，檢驗一下是否成功
            String content = getMethod.getResponseBodyAsString();
            System.out.println(content);
            System.out.println("======================4==================");

        } catch (HttpException e) {
            e.printStackTrace();
        } catch (IOException e) {
            e.printStackTrace();
        }
    }
}