Jsoup使用

最新推荐文章于 2024-08-22 11:00:00 发布

luo381821

最新推荐文章于 2024-08-22 11:00:00 发布

阅读量419

点赞数

本文链接：https://blog.csdn.net/luo381821/article/details/105305921

版权

<dependency>
     <groupId>org.jsoup</groupId>
     <artifactId>jsoup</artifactId>
     <version>1.13.1</version>
</dependency>

public class JsoupTest {

    public static void main(String[] args) throws Exception{
        String html = "<html><head><title>First parse</title></head>"
                + "<body><p>Parsed HTML into a doc.</p><p>gewgewg</p></body></html>";
        Document doc1 = Jsoup.parse(html);
        doc1.getElementsByTag("p").forEach(e->System.out.println(e.text()));

        Document doc2 = Jsoup.connect("https://www.oschina.net/search?scope=project&q=%E7%88%AC%E8%99%AB")
                .data("query", "Java")
                .userAgent("Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.132 Safari/537.36")
                .cookie("auth", "token")
                .timeout(3000)
                .get();
        doc2.getElementsByClass("project-item").forEach(e->System.out.println(e.outerHtml()));

        File input = new File("F:\\bak\\v2/websocket.html");
        Document doc3 = Jsoup.parse(input, "UTF-8");
        doc3.getElementsByTag("h1").forEach(e->System.out.println(e.outerHtml()));
    }
}