java爬虫Demo

package com.lezhi; import org.jsoup.Jsoup; import org.jsoup.nodes.Document; import org.jsoup.nodes.Element; import org.jsoup.select.Elements; public class PaChongTest { public void test() { try { String url = "http://www.52duzhe.com/2017_01/index.html"; Document doc = Jsoup.connect(url).get(); //获得文章标题 Elements main = doc.getElementsByClass("main clearfix"); Elements link = main.select("a[href]"); for (Element hr : link) { String href = hr.attr("abs:href"); Document inDoc = Jsoup.connect(href).get(); Elements inMain = inDoc.getElementsByClass("blkContainer"); Elements h1 = inMain.select("h1"); Elements artInfo = inMain.select(".artInfo"); Elements blkContainerSblkCon = inDoc.getElementsByClass("blkContainerSblkCon"); Elements p = blkContainerSblkCon.select("p"); String title = h1.text(); String author = artInfo.select("#pub_date").text(); String source = artInfo.select("#media_name").text(); String content = ""; for (Element contxt : p) { content += p.text(); } System.out.println("标题:" + title); System.out.println(author); System.out.println(source); System.out.println("内容:"+content); System.out.println(""); } // String s = title.toString(); // String text = title.text(); System.out.println("test"); // String title2 = doc.select("具体选择器内容").get(0).text(); } catch (Exception e) { e.printStackTrace(); } } public static void main(String[] args) { PaChongTest pt = new PaChongTest(); pt.test(); } }