給出一個Taglib的BODY_BUFFERED，bodyContent的例子

BODY_BUFFERED，bodyContent的例子：
馬克- to-win：馬克 java社區(qū)：防盜版實名手機尾號： 73203。
有時你標簽對應的java代碼會從數(shù)據(jù)庫或其他網(wǎng)絡渠道獲取數(shù)據(jù)。這些數(shù)據(jù)在最終返回jsp顯示之前，需要一個過濾修改的過程。比如去掉某些政治敏感的詞語或像本例一樣加入一個詞語“馬克-to-win”。這時就需要你用BODY_BUFFERED技術。顧名思義就是要把你返回jsp顯示的body先 buffer一下，放在BodyTagSupport的bodyContent里，你可以隨意修改之，最后再返回jsp，但是前提條件是在 doStartTag中的返回值要寫成EVAL_BODY_BUFFERED，這樣系統(tǒng)就會開辟bodyContent這個Buffer，供你運作。



例 1.2.6：


<%@ page contentType="text/html; charset=GBK" %>
<%@ taglib uri="/WEB-INF/tagExampleLib.tld" prefix="greeter" %>
<html>
   <body>
     <greeter:Hello>
        Now 時間 is: <%=new java.util.Date() %> <br>
     </greeter:Hello>
     結束
   </body>
</html>






<?xml version="1.0" encoding="ISO-8859-1" ?>
<!DOCTYPE taglib
     PUBLIC "-//Sun Microsystems, Inc.//DTD JSP Tag Library 1.1//EN"
     "http://java.sun.com/j2ee/dtds/web-jsptaglibrary_1_1.dtd">
<taglib>
    <tlibversion>1.0</tlibversion>
    <jspversion>1.1</jspversion>
    <shortname>mark-to-win</shortname>
    <tag>
        <name>Hello</name>
        <tagclass>com.marktowin.HelloWorldTag</tagclass>
    </tag>
</taglib>




/*順序是： 4)doStartTag
4.5) doAfterBody
5)doEndTag*/

package com.marktowin;
import javax.servlet.jsp.*;
import javax.servlet.jsp.tagext.*;
import java.io.*;

public class HelloWorldTag extends BodyTagSupport {
    int count = 2;
    public int doStartTag() {
        System.out.println("doStartTag");
        return EVAL_BODY_BUFFERED;
    }

    public int doAfterBody() {
        System.out.println("doAfterBody.");
        while (count > 0) {
            count--;
            /* EVAL_BODY_AGAIN requires that the body of the tag is evaluated. */
            return EVAL_BODY_AGAIN;
        }
        return SKIP_BODY;
    }

    public int doEndTag() {
        System.out.println("doEndTagqqq");
        /*the following try catch will be no use if you use return BodyTagSupport.EVAL_BODY_INCLUDE,
         but if you use return BodyTagSupport.EVAL_BODY_BUFFERED;, you must use this try catch,
        otherwise, there will be no output from IE, because the former won't use bodyContent. the
         former use "Evaluate body into existing out stream" */
            try{
              if(bodyContent != null) {
                System.out.println("body content qixy is not null");
        /*public JspWriter getEnclosingWriter() Get the enclosing JspWriter.
        writeOut(Writer out) Write the contents of this BodyContent into a Writer.
                 */
                System.out.println("current bodyContent is "+bodyContent.getString());
/*假如你用下句，原來的東西就沒了，但后面的馬克-to-win還是能輸出出來*/               
//                bodyContent.clearBody();
/*write的作用和append附加一樣，write沒有先清除過去body的作用*/               
                bodyContent.write("馬克-to-win");
                //bodyContent.writeOut(bodyContent.getEnclosingWriter());
/*writeOut(Writer out)
          Write the contents of this BodyContent into a Writer.
    不寫下句瀏覽器中沒有輸出       */               
                bodyContent.writeOut(pageContext.getOut());
              }
            }catch(IOException e){
              e.printStackTrace();
            }
        return EVAL_PAGE;
    }
}




運行jsp后，瀏覽器中輸出結果是：

Now 時間 is: Sun Apr 30 22:01:13 CST 2017
Now 時間 is: Sun Apr 30 22:01:13 CST 2017
Now 時間 is: Sun Apr 30 22:01:13 CST 2017
馬克-to-win 結束


運行jsp后，console中輸出結果是：

doStartTag
doAfterBody.
doAfterBody.
doAfterBody.
doEndTagqqq
body content qixy is not null
current bodyContent is
        Now 時間 is: Sun Apr 30 22:01:13 CST 2017 <br>
    
        Now 時間 is: Sun Apr 30 22:01:13 CST 2017 <br>
    
        Now 時間 is: Sun Apr 30 22:01:13 CST 2017 <br>