Jsoup学习-使用Jsoup抓取页面数据

来源：互联网发布：查询淘宝账号安全编辑：程序博客网时间：2024/05/22 07:48

1.所需包：

<dependency><groupId>org.jsoup</groupId><artifactId>jsoup</artifactId><version>1.8.3</version></dependency>

2.JAVA Code

package sf.jsoup.microcreativity;import java.io.File;import java.io.IOException;import org.jsoup.Jsoup;import org.jsoup.nodes.Document;import org.jsoup.safety.Whitelist;import org.jsoup.Connection;public class ParseHtml {public static void main(String[] args) {//案例1（简单获取body）：/*String html = "<html><head><title>First parse</title></head>"  + "<body><p>Parsed HTML into a doc.</p></body></html>";Document doc = Jsoup.parse(html);Jsoup.clean(html, new Whitelist());System.out.println(doc.body());*///案例2（获取WEB URLbody）：/*String url = "https://www.hao123.com/";Connection connection = Jsoup.connect(url);connection.data("query", "java");connection.userAgent("Mozilla");connection.cookie("auth", "token");connection.timeout(3000);try {Document doc = connection.post();System.out.println(doc.body());} catch (IOException e) {e.printStackTrace();}*///案例3（获取本地URL body）：File input = new File("d:/user/xxxxxx/桌面/in.html");try {Document doc = Jsoup.parse(input, "UTF-8");System.out.println(doc.body());} catch (IOException e) {e.printStackTrace();}}}

3.Jsoup中文文档，http://download.csdn.net/detail/icoudsoft_saas/9481597

0 0