Jsoup demo

平和雅
2023-12-01
需要一个包：jsoup-1.7.3.jar
原文地址：http://www.cnblogs.com/zyw-205520/p/3421687.html
import java.io.IOException;

import org.jsoup.Jsoup;
import org.jsoup.nodes.Document;
import org.jsoup.nodes.Element;
import org.jsoup.select.Elements;

public class JsoupTest {

    static String url="http://www.cnblogs.com/zyw-205520/archive/2012/12/20/2826402.html";
    /**
     * @param args
     * @throws Exception 
     */
    public static void main(String[] args) throws Exception {
//      BolgBody();
//      article();
        Blog();
    }

    private static void BolgBody()throws Exception{
        String html = "<html><head><title> 开源中国社区 </title></head>"
                + "<body><p> 这里是 jsoup 项目的相关文章 </p></body></html>";

        Document doc = Jsoup.parse(html);

        System.out.println(doc.body());

        //从URL直接加载HTML文档
        Document doc2 = Jsoup.connect(url).get();
        String title = doc2.body().toString();
        System.out.println(title);
    }

    //获取博客上的文章标题和连接
    public static void article(){
        Document doc = null;
        try {
            doc = Jsoup.connect("http://www.cnblogs.com/zyw-205520/").get();
            Elements listDiv = doc.getElementsByAttributeValue("class", "postTitle");
            for(Element element : listDiv){
                Elements links = element.getElementsByTag("a");
                for(Element link : links){
                    String linkHref = link.attr("href");
                    String linkText = link.text().trim();
                    System.out.println(linkHref);
                    System.out.println(linkText);
                }
            }
        } catch (IOException e) {
            // TODO Auto-generated catch block
            e.printStackTrace();
        }
    }

    //获取指定博客文章的内容
    public static void Blog(){
        Document doc = null;
        try {
            doc = Jsoup.connect("http://www.cnblogs.com/zyw-205520/archive/2012/12/20/2826402.html").get();
            Elements listDiv = doc.getElementsByAttributeValue("class", "postBody");
            for(Element element : listDiv){
                System.out.println(element.html());
            }
        } catch (IOException e) {
            // TODO Auto-generated catch block
            e.printStackTrace();
        }

    }
}
Jsoup demo

相关阅读

相关文章

相关问答
