- import java.io.UnsupportedEncodingException;
- import java.net.URLDecoder;
- import java.net.URLEncoder;
- import java.util.HashMap;
- import java.util.Map;
- import org.jsoup.Jsoup;
- import org.jsoup.nodes.Document;
- import org.jsoup.nodes.Element;
- import org.jsoup.select.Elements;
- import org.openqa.jetty.util.UrlEncoded;
- import com.steven.downmove.HttpClientUtil;
- public class Downloads {
- /**
- * @param args
- * @throws UnsupportedEncodingException
- */
- public static void main(String[] args) throws UnsupportedEncodingException {
- String keyword="开源社区";
- keyword=URLEncoder.encode(keyword,"gbk");
- //此类为httpclient封装的 就是直接获取网页内容 大家自己动手写个吧 这个就不发大家了
- HttpClientUtil hcu = new HttpClientUtil();
- String url = "http://baoku.360.cn/soft/search?kw="+keyword;
- String count = hcu.getResponseWithHttpClient(url, "gbk");
- Document doc = Jsoup.parse(count);
- Element ele = doc.select("div[class=main-list fr]").first();
- String info_data=ele.select("div[class=info-data]").first().text();
- System.out.println(info_data+"\\n");
- Elements searchpage_list = ele.select("div[id=searchpage-list] > dl");
- int temp=0;
- for (Element searchpage : searchpage_list) {
- String href = searchpage.select("dd > div[class=h2] > strong >a").attr("href");
- //下载地址
- String down = "http://q.soft.360.cn/get_download_url.php?type=download_url&soft_ids="+href.substring(href.lastIndexOf("/")+1, href.length());
- String downurl=hcu.getResponseWithHttpClient(down, "gbk");
- Document dos = Jsoup.parse(downurl);
- String urldown = dos.select("ret > softs > soft > durls").first().text();
- urldown=urldown.substring(urldown.indexOf("http"), urldown.length());
- String title = searchpage.select("dd > div[class=h2] > strong >a").first().text();
- String explain = searchpage.select("dd > p").first().text();
- String size = searchpage.select("dd > p").last().text();
- System.out.println(title+"\\n"+urldown+"\\n"+explain+"\\n"+size+"\\n");
- temp++;
- }
- System.out.println("共"+temp+"条");
- }
- }
- //该片段来自于http://www.codesnippet.cn/detail/121120137069.html
来源: http://www.codesnippet.cn/detail/121120137069.html