双色球2013年统计,从网上爬虫出来的

package com.hpu.bai;

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.File;
import java.io.FileOutputStream;
import java.io.FileWriter;
import java.io.IOException;
import java.io.Writer;
import java.net.URL;

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

public class Double2013 {
	public static void main(String[] args) throws Exception {
		Document doc;
		File file = new File("d:\双色球第2013015期到2013154期开奖结果.txt");
		FileOutputStream fos = new FileOutputStream(file); 
		//while(true){
			for(int i = 2013015;i<=2013154;i++){
			String url = "http://caipiao.163.com/award/ssq/"+i+".html";
			doc  = (Document) Jsoup.connect(url).get();
			//String title = doc.title();
			//Elements ele = doc.select("div.search_zj_left p.red_ball span");
			Element content = doc.getElementById("zj_area");
			Elements links = content.getElementsByTag("span");
			//System.out.println(links.size());
			int count =0;byte[] b = new byte[1024];
			for (Element link : links) {  
				count++;
				//String linkHref = link.attr("href");  
				String linkText = link.text();
				//System.out.print(linkHref);
				System.out.print(" "+linkText);	
				//linkText.toCharArray();
				linkText = " "+linkText;
				 b = linkText.getBytes();
				fos.write(b);
				if(count%7 == 0){
					fos.write("
".getBytes());
				}
				}
				System.out.println();
			//String ball = ele.text();
			/*Element dayFav = doc.getElementsByAttributeValue("class", "linkRight").first(); 
			Elements dayFavItems = dayFav.getElementsByTag("a"); 
			for (Element dayFavItem:dayFavItems) { 
				System.out.println(dayFavItem.text()); 
			} */
			//System.out.println(title);
			//System.out.println(ball);
			Thread.sleep(1500);
			}
		//}
	fos.flush();fos.close();
	}
}

其中用到了jsoup的东西,具体的还不太会用。QQ:623687677

原文地址:https://www.cnblogs.com/neruda/p/4025504.html