javaEE(15)_Servlet过滤器

一、Filter简介

1、Filter也称之为过滤器,它是Servlet技术中最激动人心的技术,WEB开发人员通过Filter技术,对web服务器管理的所有web资源:例如Jsp, Servlet, 静态图片文件或静态html文件等进行拦截,从而实现一些特殊的功能.例如实现网站定期自动登录、URL级别的权限访问控制、过滤敏感词汇、压缩响应信息等一些高级功能.

2、Servlet API中提供了一个Filter接口,开发web应用时,如果编写的Java类实现了这个接口,则把这个java类称之为过滤器Filter.通过Filter技术,开发人员可以实现用户在访问某个目标资源之前,对访问的请求和响应进行拦截,如下所示:

二、Filter开发入门

1、Filter开发分为二个步骤:

•编写java类实现Filter接口,并实现其doFilter方法.

•在 web.xml 文件中使用<filter>和<filter-mapping>元素对编写的filter类进行注册,并设置它所能拦截的资源.

2、Filter链

•在一个web应用中,可以开发编写多个Filter,这些Filter组合起来称之为一个Filter链.

•web服务器根据Filter在web.xml文件中的注册顺序,决定先调用哪个Filter,当第一个Filter的doFilter方法被调用时,web服务器会创建一个代表Filter链的FilterChain对象传递给该方法.在doFilter方法中,开发人员如果调用了FilterChain对象的doFilter方法,则web服务器会检查FilterChain对象中是否还有filter,如果有,则调用第2个filter,如果没有,则调用目标资源.

三、Filter的生命周期

1、init(FilterConfig filterConfig)throws ServletException:

•和我们编写的Servlet程序一样,Filter的创建和销毁由WEB服务器负责(Servlet也是). web 应用程序启动时,web 服务器将创建Filter的实例对象,并调用其init方法,完成对象的初始化功能,从而为后续的用户请求作好拦截的准备工作(注:filter对象只会创建一次(所有请求共享),init方法也只会执行一次)

•开发人员通过init方法的参数,可获得代表当前filter配置信息的FilterConfig对象.

2、destroy():

•在Web容器卸载 Filter 对象之前被调用.该方法在Filter的生命周期中仅执行一次.在这个方法中,可以释放过滤器使用的资源.

*Filter生命周期和ServletContext类似.

四、FilterConfig接口

用户在配置filter时,可以使用<init-param>为filter配置一些初始化参数,当web容器实例化Filter对象,调用其init方法时,会把封装了filter初始化参数的filterConfig对象传递进来.因此开发人员在编写filter时,通过filterConfig对象的方法,就可获得:

•String getFilterName():得到filter的名称.

•String getInitParameter(String name): 返回在部署描述中指定名称的初始化参数的值.如果不存在返回null.

•Enumeration getInitParameterNames():返回过滤器的所有初始化参数的名字的枚举集合.

•public ServletContext getServletContext():返回Servlet上下文对象的引用.

*FilterConfig和ServletConfig类似，但生命周期不同.

五、Filter案例:

1、解决全站乱码过滤器(含开发入门&Filter链&生命周期&FilterConfig)servlet3.0规范

@WebFilter(
        urlPatterns = { "/ServletDemo1" }, 
        initParams = { 
           @WebInitParam(name = "charset", value = "UTF-8", description = "编码")
        })
public class FilterDemo1 implements Filter {
    FilterConfig fConfig=null;
    public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) {
        System.out.println("FilterDemo1 doFilter 前!");
        String charset = fConfig.getInitParameter("charset");
        HttpServletRequest req = (HttpServletRequest)request;
        req.setCharacterEncoding(charset);
        
        //这两句用来设置当servlet打印页面时的编码设置,实际开发中都是转发给jsp,由jsp来通过指令和meta标签进行设置
        /*HttpServletResponse resp = (HttpServletResponse)response;
        resp.setCharacterEncoding(charset);
        resp.setContentType(charset);*/
        
        chain.doFilter(request, response);
        System.out.println("FilterDemo1 doFilter 后!");
    }

    public void init(FilterConfig fConfig) {
        System.out.println("FilterDemo1 init!");
        this.fConfig = fConfig;
    }
    
    public void destroy() {
        System.out.println("FilterDemo1 destroy!");
    }
}

@WebFilter("/ServletDemo1")
public class FilterDemo2 implements Filter {

    public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) {
        System.out.println("FilterDemo2 doFilter 前!");
        chain.doFilter(request, response);
        System.out.println("FilterDemo2 doFilter 后!");
    }

    public void init(FilterConfig fConfig) throws ServletException {
        System.out.println("FilterDemo2 init!");
    }
    
    public void destroy() {
        System.out.println("FilterDemo2 destroy!");
    }
}

@WebServlet("/ServletDemo1")
public class ServletDemo1 extends HttpServlet {

    protected void doGet(HttpServletRequest request, HttpServletResponse response) {
        String name = request.getParameter("username");
        System.out.println("ServletDemo1 doget! name="+name);
    }

    protected void doPost(HttpServletRequest request, HttpServletResponse response) {
        this.doGet(request, response);
    }
    
    public void init(ServletConfig config) throws ServletException {
        System.out.println("ServletDemo1 init!");
    }

    public void destroy() {
        System.out.println("ServletDemo1 destroy!");
    }
}

<%@ page language="java" contentType="text/html; charset=UTF-8" pageEncoding="UTF-8"%>
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Insert title here</title>
</head>
<body>
    <form action="${pageContext.request.contextPath}/ServletDemo1 " method="post">
        <input name="username" />
        <input type="submit" value="提交"/>
    </form>
</body>
</html>

运行结果:

1、当服务器启动时:

FilterDemo2 init!
FilterDemo1 init!

首先初始化FilterDemo2,与调用doFilter的顺序刚好相反(doFilter的调用顺序与类的名称相关).

2、当在demo1.jsp中输入中文点击提交时:

ServletDemo1 init!
FilterDemo1 doFilter 前!
FilterDemo2 doFilter 前!
ServletDemo1 doget! name=王维
FilterDemo2 doFilter 后!
FilterDemo1 doFilter 后!

3、当服务器关闭或者项目重新发布时:

ServletDemo1 destroy!
FilterDemo2 destroy!
FilterDemo1 destroy!

如上所示的servlet过滤器可以实现当访问说有的servlet时,在它之前进行编码设置.

2、控制浏览器缓存页面中的静态资源,禁止浏览器缓存动态页面

•场景:有些动态页面中引用了一些图片或css文件以修饰页面效果,这些图片和css文件经常是不变化的,所以为减轻服务器的压力,可以使用filter控制浏览器缓存这些文件,以提升服务器的性能.web站点优化,越在前端效果越好,比后端做缓存效果好.

<!--测试jsp页面,此页面首次访问时共会向服务器发送3次请求  -->
<%@ page language="java" contentType="text/html; charset=UTF-8" pageEncoding="UTF-8"%>
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<link rel="stylesheet" type="text/css" href="/Filter/css/1.css">
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>noCache</title>
</head>
<body>
    hello
    <img id="img1" alt="ss" src="/Filter/image/1.png">
</body>
</html>

//控制jsp不缓存的Filter,我们的jsp都是由servlet forward来的
@WebFilter(dispatcherTypes = {DispatcherType.FORWARD }, urlPatterns = { "*.jsp" })//ps备注
public class NoCacheFilter implements Filter {

    public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain){
        HttpServletResponse resp = (HttpServletResponse)response;
        
        //注意一个是setDateHeader另两个是setHeader
        resp.setDateHeader("Expires",-1);
        resp.setHeader("Cache-Control","no-cache");
        resp.setHeader("Pragma","no-cache");
        
        chain.doFilter(request, response);
    }

    public void destroy() {
    }
    public void init(FilterConfig arg0) throws ServletException {
    }
}

//控制css文件和图片缓存的Filter
@WebFilter(
        urlPatterns={"*.css","*.png"},//一个Filter可对应多个url映射,类似Servlet
        initParams = { 
             @WebInitParam(name = "cssExpires", value = "6000", description = "css文件缓存时间"),
             @WebInitParam(name = "imageExpires", value = "6000", description = "图片文件缓存时间")
        })
public class CacheFilter implements Filter {
    FilterConfig fConfig=null;
    public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain){
        HttpServletResponse resp = (HttpServletResponse)response;
        HttpServletRequest req = (HttpServletRequest)request;
        
        String cssExpires = fConfig.getInitParameter("cssExpires");
        String imageExpires = fConfig.getInitParameter("imageExpires");
        
        String uri = req.getRequestURI();
        if(uri.endsWith("css")){
            //这里注意是setDateheader不是setHeader
            resp.setDateHeader("Expires",System.currentTimeMillis()+Long.parseLong(cssExpires));
        }else{
            resp.setDateHeader("Expires",System.currentTimeMillis()+Long.parseLong(imageExpires));
        }
        chain.doFilter(request, response);
    }
    
    public void init(FilterConfig fConfig) throws ServletException {
        this.fConfig=fConfig;
    }
    public void destroy() {
    }
}

ps:1、Ctrl+F5强制刷新的话,不管浏览器有没有缓存都会向服务器重新请求.

    2、<dispatcher>指定过滤器所拦截的资源被 Servlet 容器调用的方式,可以是REQUEST,INCLUDE,FORWARD和ERROR之一,默认REQUEST.用户可以设置多个<dispatcher> 子元素用来指定 Filter 对资源的多种调用方式进行拦截.

<dispatcher> 子元素可以设置的值及其意义:
REQUEST:当用户直接访问页面时,Web容器将会调用过滤器.如果目标资源是通过RequestDispatcher的include()或forward()方法访问时,那么该过滤器就不会被调用.
INCLUDE:如果目标资源是通过RequestDispatcher的include()方法访问时,那么该过滤器将被调用.除此之外,该过滤器不会被调用.
FORWARD:如果目标资源是通过RequestDispatcher的forward()方法访问时,那么该过滤器将被调用,除此之外,该过滤器不会被调用.
ERROR:如果目标资源是通过声明式异常处理机制调用时,那么该过滤器将被调用.除此之外,过滤器不会被调用.

用法实例:

<filter-mapping>
     <filter-name>testFilter</filter-name>
     <url-pattern>/test.jsp</url-pattern>
</filter-mapping>
<filter-mapping>
    <filter-name>testFilter</filter-name>
    <url-pattern>/index.jsp</url-pattern>
    <dispatcher>FORWARD</dispatcher>
</filter-mapping>

3、实现用户自动登陆的过滤器      *代码非常经典

<!--登陆界面  -->
<%@ page language="java" import="java.util.*" pageEncoding="UTF-8"%>

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
  <head>
    <title>My JSP 'login.jsp' starting page</title>
  </head>
  
  <body>     
       <form action="${pageContext.request.contextPath }/servlet/LoginServlet" method="post">
           用户名:<input type="text" name="username"><br/>
           密码:<input type="password" name="password"><br/>
           有效期:
           1分钟<input type="radio" name="time" value="${1*60 }">
           5分钟<input type="radio" name="time" value="${5*60 }">
           10分钟<input type="radio" name="time" value="${10*60 }">
           <br/>
           <input type="submit" value="登陆">
       </form>
  </body>
</html>

//登陆Servlet
public class LoginServlet extends HttpServlet {

    public void doGet(HttpServletRequest request, HttpServletResponse response){

        String username = request.getParameter("username");
        String password = request.getParameter("password");

        BusinessService service = new BusinessService();
        User user = service.login(username, password);
        if (user == null) {
            request.setAttribute("message", "用户名或密码错误！！");
            request.getRequestDispatcher("/message.jsp").forward(request, response);
            return;
        }

        request.getSession().setAttribute("user", user);

        // 给客户机发送自动登陆的 cookie
　　　　 //把username回写给cookie,Filter需要通过username拿到用户信息,校验cookie带的md5和服务器算出的MD5是否一致,判断cookie有没有被修改
        int expirestime = Integer.parseInt(request.getParameter("time"));
        // autologin=username:expirestime:md5(password:expirestime:username)
        Cookie cookie = makeCookie(user, expirestime);
        response.addCookie(cookie);
        response.sendRedirect("/day19/index.jsp");
    }

    public Cookie makeCookie(User user, int expirestime) {
        long currenttime = System.currentTimeMillis();
        String cookieValue = user.getUsername() + ":" + (currenttime + expirestime * 1000) + ":"
                + md5(user.getUsername(), user.getPassword(), (currenttime + expirestime * 1000));
        Cookie cookie = new Cookie("autologin", cookieValue);
        cookie.setMaxAge(expirestime);
        cookie.setPath("/day19");
        return cookie;
    }

    private String md5(String username, String password, long expirestime) {
        try {
            String value = password + ":" + expirestime + ":" + username;
            MessageDigest md = MessageDigest.getInstance("md5");
            byte md5[] = md.digest(value.getBytes());
            BASE64Encoder encode = new BASE64Encoder();
            return encode.encode(md5);
        } catch (Exception e) {
            throw new RuntimeException(e);
        }
    }

    public void doPost(HttpServletRequest request, HttpServletResponse response) throws Exception {
        doGet(request, response);
    }

}

//自动登录拦截器
@WebFilter("*.*") // 访问所有资源都进行拦截
public class AutoLoginFilter implements Filter {

    public void doFilter(ServletRequest req, ServletResponse resp, FilterChain chain)
            throws IOException, ServletException {

        HttpServletRequest request = (HttpServletRequest) req;
        HttpServletResponse response = (HttpServletResponse) resp;

        // 1.先检查用户是否已登陆,没登陆才自动登陆
        User user = (User) request.getSession().getAttribute("user");
        if (user != null) {
            chain.doFilter(request, response);
            return;
        }

        // 2.没登陆,再执行自动登陆逻辑

        // 看用户有没有带自动登陆的cookie
        Cookie autoLoginCookie = null;
        Cookie cookies[] = request.getCookies();
        for (int i = 0; cookies != null && i < cookies.length; i++) {
            if (cookies[i].getName().equals("autologin")) {
                autoLoginCookie = cookies[i];
            }
        }
        if (autoLoginCookie == null) {
            chain.doFilter(request, response);
            return;
        }

        // 用户带了自动登陆的cookie,则先检查cookie的有效期
        String values[] = autoLoginCookie.getValue().split("\:");
        if (values.length != 3) {
            chain.doFilter(request, response);
            return;
        }
        long expirestime = Long.parseLong(values[1]);
        if (System.currentTimeMillis() > expirestime) {
            chain.doFilter(request, response);
            return;
        }

        // 代表cookie时间有效,再检查cookie的有效性
        String username = values[0];
        String client_md5 = values[2];

        BusinessService service = new BusinessService();
        user = service.findUser(username);
        if (user == null) {
            chain.doFilter(request, response);
            return;
        }
        //// autologin=username:expirestime:md5(password:expirestime:username)
        String server_md5 = md5(user.getUsername(), user.getPassword(), expirestime);
        if (!server_md5.equals(client_md5)) {
            chain.doFilter(request, response);
            return;
        }

        // 执行登陆
        request.getSession().setAttribute("user", user);
        chain.doFilter(request, response);
    }

    private String md5(String username, String password, long expirestime) {
        try {
            String value = password + ":" + expirestime + ":" + username;
            MessageDigest md = MessageDigest.getInstance("md5");
            byte md5[] = md.digest(value.getBytes());
            BASE64Encoder encode = new BASE64Encoder();
            return encode.encode(md5);
        } catch (Exception e) {
            throw new RuntimeException(e);
        }
    }
    public void destroy() {
    }
    public void init(FilterConfig filterConfig) throws ServletException {
    }
}

六、Decorator设计模式

•1.首先看需要被增强对象继承了什么接口或父类,编写一个类也去继承这些接口或父类.

•2.在类中定义一个变量,变量类型即需增强对象的类型.

•3.在类中定义一个构造函数,接收需增强的对象.

•4.覆盖需增强的方法,编写增强的代码.

由于开发人员在filter中可以得到代表用户请求和响应的request、response对象,因此在编程中可以使用Decorator(装饰器)模式对request、response对象进行包装,再把包装对象传给目标资源,从而实现一些特殊需求.

Servlet API 中提供了一个request对象的Decorator设计模式的默认实现类HttpServletRequestWrapper,(HttpServletRequestWrapper 类实现了request 接口中的所有方法,但这些方法的内部实现都是仅仅调用了一下所包装的的 request 对象的对应方法)以避免用户在对request对象进行增强时需要实现request接口中的所有方法.

request对象的增强案例如下:

例1、真正解决全站乱码

public class CharacterEncodingFilter2 implements Filter {

    public void doFilter(ServletRequest req, ServletResponse resp,
            FilterChain chain) throws IOException, ServletException {
        
        HttpServletRequest request = (HttpServletRequest) req;
        HttpServletResponse response = (HttpServletResponse) resp;
        
        request.setCharacterEncoding("UTF-8");  //只能解决post乱码问题
        
        chain.doFilter(new MyRequest(request), response); 
    }
    
    class MyRequest extends HttpServletRequestWrapper{

        private HttpServletRequest request;
        public MyRequest(HttpServletRequest request) {
            super(request);
            this.request = request;
        }
        @Override
        public String getParameter(String name) {
            
            String value = this.request.getParameter(name);
            if(!request.getMethod().equalsIgnoreCase("get")){
                return value;
            }
            if(value==null){
                return null;
            }
            try {
                return value = new String(value.getBytes("iso8859-1"),request.getCharacterEncoding());
            } catch (UnsupportedEncodingException e) {
                throw new RuntimeException(e);
            }
        }
    }
    public void destroy() {
    }
    public void init(FilterConfig filterConfig) throws ServletException {
    }
}

例2、使用Decorator模式包装request对象,实现html标签转义功能(Tomcat服务器中提供了转义html标签的工具类),对客户端提交的文本进行转义.

public class HtmlFilter implements Filter {

    public void doFilter(ServletRequest req, ServletResponse resp,
            FilterChain chain) throws IOException, ServletException {
        
        HttpServletRequest request = (HttpServletRequest) req;
        HttpServletResponse response = (HttpServletResponse) resp;
        
        chain.doFilter(new MyRequest(request), response);  
    }
    
    class MyRequest extends HttpServletRequestWrapper{
        private HttpServletRequest request;
        public MyRequest(HttpServletRequest request) {
            super(request);
            this.request = request;
        }
        @Override
        public String getParameter(String name) {
            
            String value = this.request.getParameter(name);
            if(value==null){
                return null;
            }
            return filter(value);
        }
        
        public String filter(String message) {

            if (message == null)
                return (null);

            char content[] = new char[message.length()];
            message.getChars(0, message.length(), content, 0);
            StringBuffer result = new StringBuffer(content.length + 50);
            for (int i = 0; i < content.length; i++) {
                switch (content[i]) {
                case '<':
                    result.append("&lt;");
                    break;
                case '>':
                    result.append("&gt;");
                    break;
                case '&':
                    result.append("&amp;");
                    break;
                case '"':
                    result.append("&quot;");
                    break;
                default:
                    result.append(content[i]);
                }
            }
            return (result.toString());
        }
    }
    public void init(FilterConfig filterConfig) throws ServletException {
    }
    public void destroy() {
    }
}

例3、词汇过滤器

其实和html标签转义功能完全类似,使用装饰器模式,拦截getParameter方法,当在servlet中调用getParameter方法时,实际调用的是我修改过的getParameter方法,该方法对词汇进行了过滤.

response对象的增强最经典案例-压缩响应如下:

1、Servlet  API 中提供了response对象的Decorator设计模式的默认实现类HttpServletResponseWrapper , (HttpServletResponseWrapper类实现了response接口中的所有方法,但这些方法的内部实现都是仅仅调用了一下所包装的的 response对象的对应方法)以避免用户在对response对象进行增强时需要实现response接口中的所有方法.

2、应用HttpServletResponseWrapper对象,压缩响应正文内容.思路:

•通过filter向目标页面传递一个自定义的response对象.

•在自定义的response对象中,重写getOutputStream方法和getWriter方法,使目标资源调用此方法输出页面内容时,获得的是我们自定义的ServletOutputStream对象.

•在我们自定义的ServletOuputStream对象中,重写write方法,使写出的数据写出到一个buffer中.

•当页面完成输出后,在filter中就可得到页面写出的数据,从而我们可以调用GzipOuputStream对数据进行压缩后再写出给浏览器,以此完成响应正文件压缩功能.

例1、没有实现全站式的压缩

//压缩输出
public class ServletDemo2 extends HttpServlet {

    public void doGet(HttpServletRequest request, HttpServletResponse response)
            throws ServletException, IOException {

        String data = "aaaaaaaa";
        
        ByteArrayOutputStream bout = new ByteArrayOutputStream();//缓存字节流
        GZIPOutputStream gout = new GZIPOutputStream(bout);
        gout.write(data.getBytes());
        gout.close();//确保写入缓存流成功
        
        byte gzip[] = bout.toByteArray();
        response.setHeader("content-encoding", "gzip");
        response.setHeader("content-length", gzip.length + "");
        
        response.getOutputStream().write(gzip);
    }

    public void doPost(HttpServletRequest request, HttpServletResponse response)
            throws ServletException, IOException {
        doGet(request, response);
    }
}

例2、全站式压缩实现

//giz压缩过滤器
public class GzipFilter implements Filter {

    public void doFilter(ServletRequest req, ServletResponse resp,
            FilterChain chain) throws IOException, ServletException {
        
        HttpServletRequest request = (HttpServletRequest) req;
        HttpServletResponse response = (HttpServletResponse) resp;
        MyResponse myresponse = new MyResponse(response);
        
        //response.getwriter  response.getOutputStream 
        chain.doFilter(request, myresponse);   
        
        //取出缓冲的数据压缩后输出
        byte out[] = myresponse.getBuffer();  //得到目标资源的输出
        System.out.println("压之前:" + out.length);
        
        byte gzipout[] = gzip(out);
        System.out.println("压之后:" + gzipout.length);
        
        response.setHeader("content-encoding", "gzip");
        response.setHeader("content-length", gzipout.length + "");
        response.getOutputStream().write(gzipout);
    }
    
    public byte[] gzip(byte b[]) throws IOException{
        
        ByteArrayOutputStream bout = new ByteArrayOutputStream();
        GZIPOutputStream gout = new GZIPOutputStream(bout);
        gout.write(b);
        gout.close();
        return bout.toByteArray();
    }
    
    class MyResponse extends HttpServletResponseWrapper{
        private ByteArrayOutputStream bout = new ByteArrayOutputStream();
        private PrintWriter pw;
        
        private HttpServletResponse response;
        public MyResponse(HttpServletResponse response) {
            super(response);
            this.response = response;
        }
        @Override
        public ServletOutputStream getOutputStream() throws IOException {
            return new MyServletOutputStream(bout);    //myresponse.getOutputStream().write("hahah");
        }
        
        @Override
        public PrintWriter getWriter() throws IOException {
            pw = new PrintWriter(new OutputStreamWriter(bout,response.getCharacterEncoding()));
            return pw;  //MyResponse.getWriter().write("中国");
        }
        public byte[] getBuffer(){
            if(pw!=null){
                pw.close();
            }
            return bout.toByteArray();
        }
    }
    
    class MyServletOutputStream extends ServletOutputStream{

        private ByteArrayOutputStream bout;
        public MyServletOutputStream(ByteArrayOutputStream bout){
            this.bout = bout;
        }
        @Override
        public void write(int b) throws IOException {
            bout.write(b);
        }
    }
    public void destroy() {
    }
    public void init(FilterConfig filterConfig) throws ServletException {
    }
}

public class GzipServlet extends HttpServlet {

    public void doGet(HttpServletRequest request, HttpServletResponse response)
            throws ServletException, IOException {

        String data = "中国";
        response.getOutputStream().write(data.getBytes("UTF-8"));
        request.getRequestDispatcher("/index.jsp").forward(request, response);
    }

    public void doPost(HttpServletRequest request, HttpServletResponse response)
            throws ServletException, IOException {
        doGet(request, response);
    }
}

web.xml配置

<filter>
    <filter-name>GzipFilter</filter-name>
    <filter-class>cn.itcast.web.filter.example.GzipFilter</filter-class>
</filter>

<filter-mapping>
    <filter-name>GzipFilter</filter-name>
    <url-pattern>*.jsp</url-pattern>
    <dispatcher>FORWARD</dispatcher>
    <dispatcher>REQUEST</dispatcher>
</filter-mapping>

<filter-mapping>
    <filter-name>GzipFilter</filter-name>
    <url-pattern>*.html</url-pattern>
</filter-mapping>

<filter-mapping>
    <filter-name>GzipFilter</filter-name>
    <url-pattern>*.js</url-pattern>
</filter-mapping>

ps:gzip一般压缩文本类型的数据,对图片视频等压缩率较低.

***有一点要注意,当servlet中以forward方式转发给另一个servlet时,实际是通知tomcat去调用另一个servlet,这其中如果有拦截器的话会执行.

七、实用案例－缓存数据到内存  

对于页面中很少更新的数据,例如商品分类,为避免每次都要从数据库查询分类数据,因此可把分类数据缓存在内存或文件中,以此来减轻数据库压力,提高系统响应速度.

//可以配置给几个指定的servlet
public class WebCacheFilter implements Filter {
    
    private Map<String,byte[]> map = new HashMap();
    
    public void doFilter(ServletRequest req, ServletResponse resp,
            FilterChain chain) throws IOException, ServletException {
        
        HttpServletRequest request = (HttpServletRequest) req;
        HttpServletResponse response = (HttpServletResponse) resp;
        
        //1.得到用户想访问的资源(uri)
        String uri = request.getRequestURI();
        
        //2.看map集合中是否保存了该资源的数据
        byte b[] = map.get(uri);
        
        //3.如果保存了,则直接取数据打给浏览器
        if(b!=null){
            response.getOutputStream().write(b);
            return;//servlet就不执行了
        }
        
        //4.如果没有保存数据,则放行让目标资源执行,这时还需写一个response的包装类,捕获目标资源的输出
        MyResponse my = new MyResponse(response);
        chain.doFilter(request, my);
        byte data[] = my.getBuffer();
        
        //5.以资源uri为关键字,打资源的数据保存map集合中,以备于下次访问
        map.put(uri, data);
        
        //6.输出数据给浏览器
        response.getOutputStream().write(data);
    }
    
    class MyResponse extends HttpServletResponseWrapper{
        private ByteArrayOutputStream bout = new ByteArrayOutputStream();
        private PrintWriter pw;
        
        private HttpServletResponse response;
        public MyResponse(HttpServletResponse response) {
            super(response);
            this.response = response;
        }
        @Override
        public ServletOutputStream getOutputStream() throws IOException {
            return new MyServletOutputStream(bout);    //myresponse.getOutputStream().write("hahah");
        }
        
        @Override
        public PrintWriter getWriter() throws IOException {
            pw = new PrintWriter(new OutputStreamWriter(bout,response.getCharacterEncoding()));
            return pw;  //MyResponse.getWriter().write("中国");
        }
        public byte[] getBuffer(){
            if(pw!=null){
                pw.close();
            }
            return bout.toByteArray();
        }
    }
    
    class MyServletOutputStream extends ServletOutputStream{

        private ByteArrayOutputStream bout;
        public MyServletOutputStream(ByteArrayOutputStream bout){
            this.bout = bout;
        }
        @Override
        public void write(int b) throws IOException {
            bout.write(b);
        }
    }
    public void init(FilterConfig filterConfig) throws ServletException {
    }
    public void destroy() {
    }
}

ps:过滤器经典案例，权限系统，略。