jsoup获取文档类示例

 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 WyCrawler {
     public static void main(String[] args) {
         try {
             Document document = Jsoup.connect("http://某网页").timeout(3000).get();
             String selector = "li>div[class=titleBar clearfix]>h3>a";
             Elements elements = document.select(selector);
             for(Element element:elements){
 //                System.out.println(element.text());
                 String url = element.absUrl("href");
                 Document document2 = Jsoup.connect(url).get();
                 Elements elements2 = document2.select("#endText");
                 for(Element element2:elements2){
                     System.out.println(element2.text());
                 }
             }
         } catch (IOException e) {
             e.printStackTrace();
         }
     }
 }

上面是如何爬取超链接里的内容，下面的比较简单

 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 Test {
     public static void main(String[] args) {
         try {
             Document document = Jsoup.connect("http://www.某网页.com/").get();
             //获取内容
 //            String selector = "div[class=panel panel20 post-item post-box]>div[class=item-detail]>div[class=item-content]";
 //            Elements elements = document.select(selector);
 //            for(Element element:elements){
 //                System.out.println(element.text());
 //            }

             //获取标题
 //            String selector2 = "div[class=panel panel20 post-item post-box]>div[class=item-detail]>h2[class=item-title]";
 //            Elements elements = document.select(selector2);
 //            for(Element element:elements){
 //                System.out.println(element.text());
 //            }

             //综合写法，标题内容一起获取
             String selector = "div[class=panel panel20 post-item post-box]>div[class=item-detail]";
             Elements elements = document.select(selector);
             for(Element element:elements){
                 Elements titles = element.select("div[class=item-title]");
                 Elements content = element.select("h2[class=item-content]");
                 System.out.println(titles.text()+"\n"+content.text());
             }

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