正则表达式的威力5_网页爬虫，爬邮件地址

/*
正则表达式-网页爬虫（蜘蛛）,爬邮箱
*/

import java.io.*;
import java.util.regex.*;
import java.net.*;

class  RegexTest2
{
	public static void main(String[] args) throws Exception
	{
		//getMails();
		getMails_1();
	}

	//获取网页中的邮件地址
	public static void getMails_1() throws Exception
	{	
		URL url=new URL("http://168.245.124.92:8080/myweb/mail.html");
		URLConnection conn=url.openConnection();

		BufferedReader bufIn=new BufferedReader(new InputStreamReader(conn.getInputStream()));

		String line=null;

		String reg="\\w+@\\w+(\\.\\w+)+";
		Pattern p=Pattern.compile(reg);
		
		while ((line=bufIn.readLine())!=null)
		{
			Matcher m=p.matcher(line);
			while (m.find())
			{
				System.out.println(m.group());
			}
		}		
	}

	/*获取指定文档中的邮件地址
	使用获取功能,Pattern,Matcher
	*/
	public static void getMails() throws Exception
	{
		BufferedReader bufr=new BufferedReader(new FileReader("mails.txt"));

		String line=null;

		String reg="\\w+@\\w+(\\.\\w+)+";
		Pattern p=Pattern.compile(reg);
		
		while ((line=bufr.readLine())!=null)
		{
			Matcher m=p.matcher(line);
			while (m.find())
			{
				System.out.println(m.group());
			}
		}
	}
}