链家爬取深圳租房代码(java)

jsoup的jar包maven导入

<dependencies>
<!--        爬虫-->
        <dependency>
            <groupId>org.jsoup</groupId>
            <artifactId>jsoup</artifactId>
            <version>1.13.1</version>
        </dependency>
</dependencies>

爬取的内容打印到控制台
你也可以输出到文件
在输出到文件方法中的注释就是输出到文件，你修改一下文件路径和文件的格式就可以了。

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

import java.io.*;
import java.util.HashMap;
import java.util.Map;


/**
 * @author kliker
 * @version 1.0
 */
//生成文档注释 javadoc -d F:\SSM\data -author -version Main.java
public class Main {
    private static Map<String,String> map = new HashMap<>();
    public static void main(String[] args) {
        for(int i = 1;i<101;i++) {
            String url = "https://sz.lianjia.com/zufang/pg"+i+"rt200600000002/#contentList";
            try {
                Document html = Jsoup.connect(url).get();
                Elements elements = html.getElementsByClass("twoline");
                for (Element element : elements) {
                    String title = element.text();
                    String content = element.getElementsByTag("a").attr("href");
//                    System.out.println(title + "\t" + "https://sz.lianjia.com" + content);
                    String contents = getHtml(content);
                    map.put(title,contents);
                }
                writeFile(map);
            } catch (IOException e) {
                e.printStackTrace();
            } catch (Exception e) {
                e.printStackTrace();
            }
        }
    }
    //获取房子详细信息
    public static String getHtml(String content){
        String url = "https://sz.lianjia.com"+content;
        String contents1 = null;
        String contents2 = null;
        try {
            Document html = Jsoup.connect(url).get();
            Elements elements1 = html.select("#info");
//            Elements elements2 = html.select("body > div.wrapper > div:nth-child(2) > div.content.clear.w1150 > div.content__detail > div.content__article.fl > ul");
//            Elements elementsImg = html.select("#gslide > div > p.gslide--list");
//            for(Element element : elementsImg){
//                String attr = element.getElementsByTag("img").attr("src");
//                getImg(attr);
//            }
            contents1 = elements1.text();
//            contents2 = elements2.text();
            System.out.println(contents1);
        }catch(Exception e){
            e.printStackTrace();
        }
        return contents1;
    }
    //获取房子图片
//    public static void getImg(String urlImg){
//        try {
//            Document html = Jsoup.connect(urlImg).get();
//        }catch(Exception e){
//            e.printStackTrace();
//        }
//    }
    //写入文件
    public static void writeFile(Map<String,String> map) {
        try {
//            BufferedWriter bf = new BufferedWriter(new OutputStreamWriter(new FileOutputStream(new File("data/shenzhenzufang.txt"))));
//            for(Map.Entry<String,String> entry:map.entrySet()) {
//                bf.write(entry.getKey());
//                bf.newLine();
//                bf.write(entry.getValue());
//                bf.newLine();
//            }
//            bf.flush();
//            bf.close();
            PrintWriter pw = new PrintWriter(new OutputStreamWriter(System.out));
            for(Map.Entry<String,String> entry:map.entrySet()) {
                pw.println(entry.getKey());
                pw.println(entry.getValue());
            }
        }catch(Exception e){
            e.printStackTrace();
        }
    }
}