2011.06.29——— Jsoup HttpClient 抓取网络上的图片
2011.06.29——— Jsoup HttpClient 抓取网络上的图片

参考：http://www.iteye.com/topic/1106648
http://www.ibm.com/developerworks/cn/java/j-lo-jsouphtml/index.html?ca=drs-

jsoup 官方网站：http://jsoup.org

需要的主要jar包
httpclient-4.0.1jar jsoup-1.5.2.jar

主要代码 如下

Exmaple3.java

package com.th.spider.test;


import java.io.BufferedOutputStream;
import java.io.FileOutputStream;

import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.apache.http.HttpEntity;
import org.apache.http.HttpResponse;
import org.apache.http.HttpStatus;
import org.apache.http.client.HttpClient;
import org.apache.http.client.methods.HttpGet;
import org.apache.http.impl.client.DefaultHttpClient;
import org.apache.http.params.CoreConnectionPNames;
import org.apache.http.util.EntityUtils;
import org.jsoup.Connection;
import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;

public class Exmaple3 {

	private static final Log log = LogFactory.getLog(Exmaple3.class);
	
	/**
	 * 抓取图片存放目录
	 */
	private static final String PIC_DIR = "/home/li/pic";
	
	/**
	 * 链接超时
	 */
	private static final int TIME_OUT = 5000;
	
	static void go3(String url) throws Exception {
        Connection conn= Jsoup.connect(url);
        Document doc = conn.get();
        Elements links = doc.select("div.piclist img[src]");
        for(int i=0;i<links.size();i++){
            Element element = links.get(i);
            final String imgUrl = element.attr("src");
            log.info(imgUrl);
            Thread.sleep(500);
            new Thread(new Runnable() {
                public void run() {
                    try {
                        save(imgUrl);
                    } catch (Exception e) {
                        // TODO Auto-generated catch block
                        e.printStackTrace();
                    }
                }
            }).start();
        }
    }
	
	static void go2(String url) throws Exception {
	    Connection conn= Jsoup.connect(url);
	    Document doc = conn.get();
	    Elements links = doc.select("div.cc a[href]");
	    for(int i=0;i<links.size();i++){
	        Element element = links.get(i);
	        final String dirUrl = "http://www.3lian.com"+element.attr("href");
	        log.info(dirUrl);
	        Thread.sleep(500);
            new Thread(new Runnable() {
                public void run() {
                    try {
                        Connection conn= Jsoup.connect(dirUrl);
                        Document doc = conn.get();
                        Elements images = doc.select("div.mb_jjnr img[src]");
                        for(int j=0;j<images.size();j++){
                            Element img = images.get(j);
                            String imgUrl = img.attr("src");
                            log.info(imgUrl);
                            save(imgUrl);
                        }
                    } catch (Exception e) {
                        e.printStackTrace();
                    }
                }
            }).start();
	    }
    }
	
	/**
	 * 处理帖子URL
	 * @param url
	 * @throws Exception
	 */
	static void go(String url) throws Exception {
		// JSOP创建链接
		Connection conn = Jsoup.connect(url);
		// 请求返回整个文档对象
		Document doc = conn.post();
		// 选择所有class=zoom 的img标签对象
		Elements imgs = doc.select("img[class=zoom]");
		// 循环每个img标签
		for (int i = 0; i < imgs.size(); i++) {
			Element img = imgs.get(i);
			// 取得图片的下载地址
			String picURL = doc.baseUri() + img.attr("file");
			log.info(picURL);
			// 保存图片
			save(picURL);
		}
	}
	
	//<img src="static/image/common/none.gif" file="data/attachment/forum/201105/08/174412nz3jq4z90s33s2t0.jpg" width="770" class="zoom" onclick="zoom(this, this.src)" id="aimg_180565" onmouseover="showMenu({'ctrlid':this.id,'pos':'12'})" alt="img_src_29620.jpg" title