第一次寫爬蟲嗦枢，案例比較簡(jiǎn)單，就在智聯(lián)招聘網(wǎng)站上爬取職位信息屯断。

技術(shù)點(diǎn):IO流文虏，集合，Jsoup使用殖演，以及前端的知識(shí)

這里寫圖片描述

代碼如下:

package com.wty.utils;

import java.io.BufferedReader;

import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.net.URL;
import java.net.URLConnection;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;

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

/**
 * 
 *<p>Title:智聯(lián)招聘網(wǎng)站爬蟲小案例</p>
 *<p>Description:</p>
 *<p>Company:</p>
 *@author wty
 *@date 2017年3月11日上午11:28:01
 *
 */
public class DownWebPageUtils {
    public static String getHtmlCode(String url, String encoding) {
        URL uri =null;
        URLConnection urlConnection =null;
        InputStream inputStream =null;
        InputStreamReader inputStreamReader = null;
        BufferedReader bReader =null;
        StringBuffer sBuffer= new StringBuffer();
    
        try {
            // 建立網(wǎng)絡(luò)連接
             uri = new URL(url);
            // 打開(kāi)連接
             urlConnection = uri.openConnection();
            //輸入流
             inputStream = urlConnection.getInputStream();
            inputStreamReader = new InputStreamReader(inputStream, encoding);
            bReader = new BufferedReader(inputStreamReader);
            String temp;
            while ((temp = bReader.readLine()) != null) {
                sBuffer.append(temp + "\n");
            }

        } catch (Exception e) {
            e.printStackTrace();
            
        }finally {
            //關(guān)閉資源
            if(bReader!=null){
                try {
                    bReader.close();
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }
        }
        return sBuffer.toString();
}
    public static List<HashMap<String, String>> analyzeHtml(String url, String encoding){
        String htmlCode = getHtmlCode(url, encoding);
        Document document = Jsoup.parse(htmlCode);
        Elements elements = document.getElementsByClass("newlist");
        List<HashMap<String, String>> list=new ArrayList<>();
        for (Element e : elements) {
            HashMap<String, String> map = new HashMap<>();
            String zwmc = e.getElementsByClass("zwmc").text();
            String gsmc = e.getElementsByClass("gsmc").text();
            String zwyx = e.getElementsByClass("zwyx").text();
            String gzdd = e.getElementsByClass("gzdd").text();
            String gxsj = e.getElementsByClass("gxsj").text();
            map.put("職位名稱:", zwmc);
            map.put("公司名稱", gsmc);
            map.put("職位月薪", zwyx);
            map.put("工作地點(diǎn)", gzdd);
            map.put("發(fā)布日期", gxsj);
            list.add(map);
        }
        
        return list;
    }
    public static void main(String[] args) {
        List<HashMap<String, String>> resultList = analyzeHtml("http://sou.zhaopin.com/jobs/searchresult.ashx?jl=%E4%B8%8A%E6%B5%B7&kw=java&sm=0&p=1", "UTF-8");
        System.out.println(resultList);
    }
}

數(shù)據(jù)可以在控制臺(tái)中打印出來(lái)氧秘，至于jsp頁(yè)面的展現(xiàn)筆者還沒(méi)寫，無(wú)非就是表格趴久，循環(huán)什么的丸相。有興趣的話可以寫一下。

感覺(jué)爬蟲還蠻好玩的彼棍。下次準(zhǔn)備爬個(gè)美圖網(wǎng)站灭忠。。座硕。hhhhh