day25/RegexTest2.java

/*
网页爬虫(蜘蛛)
 
*/

import java.io.*;
import java.util.regex.*;
import java.net.*;
class RegexTest2 
{
	public static void main(String[] args) throws Exception
	{
		//getMail();
		getTomcatMail();
	}


	//获取指定文件中的邮箱地址
	//使用获取功能 Pattern Matcher
	public static void getMail()throws Exception
	{
		BufferedReader bufr = new BufferedReader(new FileReader("mail.txt"));		
		
		String line=null;
		Pattern p = Pattern.compile("[a-zA-Z0-9_]+@[a-z]+(\\.[a-z]+)+");
		while((line=bufr.readLine())!=null)
		{
			Matcher  m = p.matcher(line);
			while(m.find())
			{
				System.out.println(m.group());
			}
		}
	}

	//获取Tomcat服务器上网页mail.html内容中的邮箱
	public static void getTomcatMail() throws Exception
	{
		//URL url = new URL("http://localhost:8080/myweb/mail.html");
		URL url = new URL("http://zhidao.baidu.com/link?url=xF77TDRnMmNWlPTMJIdwcZn4CEHrKjhJBpCPXh23Dwxm-wGMKZXPkA8JkwmIXApGJBHX5xZJOhXa9jQRX9PkI_");

		URLConnection conn = url.openConnection();

		InputStream in = conn.getInputStream();

		BufferedReader bufrIn = new BufferedReader(new InputStreamReader(in));

		Pattern p = Pattern.compile("[a-zA-Z0-9_]+@[a-z0-9]+(\\.[a-z]+)+");

		String line = null;
		while((line=bufrIn.readLine())!=null)
		{
			Matcher m = p.matcher(line);
			while(m.find())
			{
				System.out.println(m.group());
			}
		}
	}
}