使用gzip优化web应用(filter实现)

2009-05-18
使用gzip优化web应用(filter实现)


以前使用过filter,也就是屏蔽一下受限制的资源访问路径,解决下中文乱码问题,其实filter在优化web应用发面也有出色的应用,我们可以使用filter,结合gzip 压缩技术,解决web应用中网络传输数据量大的问题,一般使用了gzip压缩,网络的传输流量能减少40%作用,效果还是相当明显的.在工作中,gzip在企业级的应用中还不是很普遍,也许是程序员一开始想的就是sql优化,缓存等更直接有效的方式,而忽略了gzip压缩方法吧.





为什么要开启Gzip ?

gzip是http协议中使用的一种加密算法,客户端向web服务器端发出了请求后，通常情况下服务器端会将页面文件和其他资源，返回到客户端，客户端加载后渲染呈现，这种情况文件一般都比较大，如果开启Gzip ，那么服务器端响应后，会将页面，JS,CSS等文本文件或者其他文件通过高压缩算法将其压缩，然后传输到客户端，由客户端的浏览器负责解压缩与呈现。通常能节省40%以上的流量（一般都有60%左右），一些PHP，JSP文件也能够进行压缩。




那怎么开启呢？

Tomcat 开启Gzip ：

1.找到Tomcat 目录下的conf下的server.xml，并找到如下信息

Connector port="8080" maxHttpHeaderSize="8192" maxThreads="150" minSpareThreads="25" maxSpareThreads="75" enableLookups="false" redirectPort="8443" acceptCount="100" connectionTimeout="20000" disableUploadTimeout="true"
将它改成如下的形式（其实在上面代码的下面已经有了，将他们打开而已。）：

<!-- Define a non-SSL HTTP/1.1 Connector on port 8080 --> <Connector port="8080" maxHttpHeaderSize="8192" maxThreads="150" minSpareThreads="25" maxSpareThreads="75" enableLookups="false" redirectPort="8443" acceptCount="100" connectionTimeout="20000" disableUploadTimeout="true" compression="on" compressionMinSize="2048" noCompressionUserAgents="gozilla, traviata" compressableMimeType="text/html,text/xml" >
这样，就能够对html和xml进行压缩了，如果要压缩css 和 js，那么需要将

compressableMimeType=”text/html,text/xml”加入css和js：

<Connector port="8080" ......... compressableMimeType="text/html,text/xml,text/css,text/javascript" >
你甚至可以压缩图片：

compressableMimeType=”text/html,text/xml”加入css和js：

<Connector port="8080" ......... compressableMimeType="text/html,text/xml,text/css,text/javascript,image/gif,image/jpg" >


——开启后——–

开启后重启Tomcat ，通过浏览器查看headers信息就能看到是否开启（firebug中有），如果开启了，那么transfer-encoding就会是Gzip，否则就是chunked。



以上方式依赖了tomcat服务器的特性,如果你服务器是iis或者websphere那么还有不同的配置方式,下面我们使用filter,在代码级别完成web应用的gzip压缩的开启.



实现定制输出的关键是对HttpServletResponse 进行包装，截获所有的输出，等到过滤器链处理完毕后，再对截获的输出进行处理，并写入到真正的HttpServletResponse 对象中。JavaEE 框架已经定义了一个HttpServletResponseWrapper 类使得包装HttpServletResponse 更加容易。我们扩展这个HttpServletResponseWrapper，截获所有的输出，并保存到ByteArrayOutputStream 中



步骤:







1.Wrapper 用来包装HttpServletResponse 对象





Java代码
1.public class Wrapper extends HttpServletResponseWrapper
2.{
3. public static final int OT_NONE = 0, OT_WRITER = 1, OT_STREAM = 2;
4. private int outputType = OT_NONE;
5. private ServletOutputStream output = null;
6. private PrintWriter writer = null;
7. private ByteArrayOutputStream buffer = null;
8. public Wrapper(HttpServletResponse resp) throws IOException {
9. super(resp);
10. buffer = new ByteArrayOutputStream();
11. }
12. public PrintWriter getWriter() throws IOException {
13. if(outputType==OT_STREAM)
14. throw new IllegalStateException();
15. else if(outputType==OT_WRITER)
16. return writer;
17. else {
18. outputType = OT_WRITER;
19. writer = new PrintWriter(new OutputStreamWriter(buffer, getCharacterEncoding()));
20. return writer;
21. }
22. }
23. public ServletOutputStream getOutputStream() throws IOException {
24. if(outputType==OT_WRITER)
25. throw new IllegalStateException();
26. else if(outputType==OT_STREAM)
27. return output;
28. else {
29. outputType = OT_STREAM;
30. output = new WrappedOutputStream(buffer);
31. return output;
32. }
33. }
34. public void flushBuffer() throws IOException {
35. if(outputType==OT_WRITER)
36. writer.flush();
37. if(outputType==OT_STREAM)
38. output.flush();
39. }
40. public void reset() {
41. outputType = OT_NONE;
42. buffer.reset();
43. }
44. public byte[] getResponseData() throws IOException {
45. flushBuffer();
46. return buffer.toByteArray();
47.
48. }
49. class WrappedOutputStream extends ServletOutputStream {
50. private ByteArrayOutputStream buffer;
51. public WrappedOutputStream(ByteArrayOutputStream buffer) {
52. this.buffer = buffer;
53. }
54. public void write(int b) throws IOException {
55. buffer.write(b);
56. }
57. public byte[] toByteArray() {
58. return buffer.toByteArray();
59. }
60. }
61.}
public class Wrapper extends HttpServletResponseWrapper
{
public static final int OT_NONE = 0, OT_WRITER = 1, OT_STREAM = 2;
private int outputType = OT_NONE;
private ServletOutputStream output = null;
private PrintWriter writer = null;
private ByteArrayOutputStream buffer = null;
public Wrapper(HttpServletResponse resp) throws IOException {
super(resp);
buffer = new ByteArrayOutputStream();
}
public PrintWriter getWriter() throws IOException {
if(outputType==OT_STREAM)
throw new IllegalStateException();
else if(outputType==OT_WRITER)
return writer;
else {
outputType = OT_WRITER;
writer = new PrintWriter(new OutputStreamWriter(buffer, getCharacterEncoding()));
return writer;
}
}
public ServletOutputStream getOutputStream() throws IOException {
if(outputType==OT_WRITER)
throw new IllegalStateException();
else if(outputType==OT_STREAM)
return output;
else {
outputType = OT_STREAM;
output = new WrappedOutputStream(buffer);
return output;
}
}
public void flushBuffer() throws IOException {
if(outputType==OT_WRITER)
writer.flush();
if(outputType==OT_STREAM)
output.flush();
}
public void reset() {
outputType = OT_NONE;
buffer.reset();
}
public byte[] getResponseData() throws IOException {
flushBuffer();
return buffer.toByteArray();

}
class WrappedOutputStream extends ServletOutputStream {
private ByteArrayOutputStream buffer;
public WrappedOutputStream(ByteArrayOutputStream buffer) {
this.buffer = buffer;
}
public void write(int b) throws IOException {
buffer.write(b);
}
public byte[] toByteArray() {
return buffer.toByteArray();
}
}
}






2.过滤器



Java代码
1.public class GZipFilter implements Filter
2.{
3.
4. public void destroy()
5. {
6. // TODO Auto-generated method stub
7.
8. }
9.
10. public void doFilter(ServletRequest request, ServletResponse response,
11. FilterChain chain) throws IOException, ServletException
12. {
13. System.out.println("进入过滤器");
14. HttpServletResponse resp = (HttpServletResponse)response;
15. Wrapper wrapper = new Wrapper(resp);
16. chain.doFilter(request, wrapper);
17. byte[] gzipData = gzip(wrapper.getResponseData());
18. resp.addHeader("Content-Encoding", "gzip");
19. resp.setContentLength(gzipData.length);
20. ServletOutputStream output = response.getOutputStream();
21. output.write(gzipData);
22. output.flush();
23. }
24.
25. public void init(FilterConfig arg0) throws ServletException
26. {
27. // TODO Auto-generated method stub
28.
29. }
30.
31. private byte[] gzip(byte[] data) {
32. ByteArrayOutputStream byteOutput = new ByteArrayOutputStream(10240);
33. GZIPOutputStream output = null;
34. try {
35. output = new GZIPOutputStream(byteOutput);
36. output.write(data);
37. }
38. catch (IOException e) {}
39. finally {
40. try {
41. output.close();
42. }
43. catch (IOException e) {}
44. }
45. return byteOutput.toByteArray();
46. }
47.
48.}
public class GZipFilter implements Filter
{

public void destroy()
{
// TODO Auto-generated method stub

}

public void doFilter(ServletRequest request, ServletResponse response,
FilterChain chain) throws IOException, ServletException
{
System.out.println("进入过滤器");
HttpServletResponse resp = (HttpServletResponse)response;
Wrapper wrapper = new Wrapper(resp);
chain.doFilter(request, wrapper);
byte[] gzipData = gzip(wrapper.getResponseData());
resp.addHeader("Content-Encoding", "gzip");
resp.setContentLength(gzipData.length);
ServletOutputStream output = response.getOutputStream();
output.write(gzipData);
output.flush();
}

public void init(FilterConfig arg0) throws ServletException
{
// TODO Auto-generated method stub

}

private byte[] gzip(byte[] data) {
ByteArrayOutputStream byteOutput = new ByteArrayOutputStream(10240);
GZIPOutputStream output = null;
try {
output = new GZIPOutputStream(byteOutput);
output.write(data);
}
catch (IOException e) {}
finally {
try {
output.close();
}
catch (IOException e) {}
}
return byteOutput.toByteArray();
}

}











3.在web.xml中配置 GZipFilter,当我们访问应用中以.html结尾的资源的使用,服务器端就开启http gzip压缩,将压缩后的信息通过http 协议传递给浏览器.





Xml代码
1.<filter>
2. <filter-name>ecsideExport</filter-name>
3. <filter-class>cn.com.xinli.test.GZipFilter</filter-class>
4. </filter>
5. <filter-mapping>
6. <filter-name>ecsideExport</filter-name>
7. <url-pattern>*.html</url-pattern>
8. </filter-mapping>
9.
<filter>
<filter-name>ecsideExport</filter-name>
<filter-class>cn.com.xinli.test.GZipFilter</filter-class>
</filter>
<filter-mapping>
<filter-name>ecsideExport</filter-name>
<url-pattern>*.html</url-pattern>
</filter-mapping>




测试: 首先下载ieHttpHeader 插件.



测试地址:

http://localhost:9090/dwr/test.html



1.未开启http gzip 压缩



GET /dwr/test.html HTTP/1.1
Accept: */*
Accept-Language: zh-cn
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1)
Host: localhost:9090
Connection: Keep-Alive


HTTP/1.1 200 OK
Server: Apache-Coyote/1.1
ETag: W/"5060-1242444154000"
Last-Modified: Sat, 16 May 2009 03:22:34 GMT
Content-Type: text/html
Content-Length: 5060
Date: Mon, 18 May 2009 12:29:49 GMT





2.开启http gzip 压缩



GET /dwr/test.html HTTP/1.1
Accept: image/gif, image/x-xbitmap, image/jpeg, image/pjpeg, application/x-shockwave-flash, application/vnd.ms-excel, application/vnd.ms-powerpoint, application/msword, */*
Accept-Language: zh-cn
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1)
Host: localhost:9090
Connection: Keep-Alive

HTTP/1.1 200 OK
Server: Apache-Coyote/1.1
ETag: W/"5060-1242444154000"
Last-Modified: Sat, 16 May 2009 03:22:34 GMT
Content-Encoding: gzip
Content-Type: text/html
Content-Length: 837
Date: Mon, 18 May 2009 12:27:33 GMT



效果是非常明显的..........