How to use XML to implement the structure of a multi-channel access website

1. Background

In today’s websites, there are more and more access channels, and the technology is becoming more and more advanced, such as WAP, SMS, EMAIL, traditional Web, Socket, etc. If even the database and LDAP are considered to be connected, then the space that needs to be expanded in the design must be very good to ensure that when adding new channels, no more code modifications or even code changes are required. . But is it possible? I don’t think it is possible, but is there any way to better solve the perfection of this multi-channel access framework?

2. Architecture

[Figure 1]

As shown in Figure 1, when all the existing accesses have been used, the designers are dazzled. If it is to get a share, You can write these programs in any way, and they can definitely be implemented, but it will be more painful to maintain. Let’s go back to the question, how can we achieve a more perfect implementation? Figure 2 shows:

[Picture 2]

Picture 2 looks like an eight-clawed octopus. The octopus legs are connected to all access channels respectively. The core of connecting all these channels is the octopus head xmlRouter. The role of the Router here is to communicate across all channels, implement data routing, and enhance the scalability and flexibility of the system's architecture. The benefits will be many. It is called XMLRouter because if XML, a flexible and standardized language, is not used as a data transmission medium, the workload of the Router will also increase exponentially. Defining the XML specification will bring about future expansion. Comes with many benefits.

3. Ideas and Patterns

The original idea of ​​XMLRouter came from the Builder Pattern in the computer motherboard and . The PCI slot of the computer motherboard defines the PCI specification. As long as The card you produce complies with the PCI standard, then you can plug it into the motherboard and it will work. As for how it works inside, it has been encapsulated. The Builder Pattern proposes to separate complex construction and implement it step by step. .XMLRouter separates these complex channels and performs them one by one.

Services idea: In order to communicate with Router, a unified interface must be defined when these channels are connected, which is called Services. As long as the program conforms to the Services specification, it can access the Router and route data.

Factory mode and Composite mode

XMLRouter will be generated using the Factory mode in the actual design, and the Router is generated by the RouterFactory Production will be placed in the queue when it is put into use. The corresponding Router for transmitting data, receiving data and returning data is called from the queue, and the Composite mode is applied.

IV. XML configuration file

The XML file is divided into two parts for use in Router. The first one is the configuration of Router, such as:

The following is a reference fragment:

<?xml version="1.0" ?> 
<services> 
　　 <!-- database Service --> 
　　 <service name="database" type="database" class="com.web.service.DBService"> 
　　 <connector 
　　　　　　 driver="com.microsoft.jdbc.sqlserver.SQLServerDriver" 
　　　　　　 url="jdbc:microsoft:sqlserver://192.168.0.179:1433" user="test" 
　　　　　　 passwd="test" /> 
　　 </service> 
　　 <!-- Web Service--> 
　　 <service name="web" type="web" class="com.web.service.WebService" > 
　　　　 <connector　/> 
　　 </service> 
…… 
</services>

This is the configuration file of Router. The service node represents the channel that needs to be accessed. The service node contains connector sub-nodes. The configuration of the sub-nodes is distinguished according to type. If it is database, then Contains attributes such as url, user, passwd, driver, etc. If it is a socket, it includes attributes such as port, maxthread, etc. The attribute values ​​​​can be configured according to your own definition.

Another XML file is the XML transaction data file. Used to transfer data in all services. Each Service contains a corresponding XML file. For example, the format of webtrans.xml is as follows:

The following is a reference fragment:

<?xml version="1.0" ?> 
<transaction> 
　　 <trans name="addDoc" service="database" method="insert"> 
　　　　 <PRoperty name="createtime" type="timestamp"/> 
　　　　 <property name="creatorid" type="long"/> 
　　　　 <property name="doctypeid" type="int"/> 
　　　　 <property name="docstatusid" type="int"/>　　　　　　 
　　 </trans>　　　　 
</transaction>

The corresponding dbtrans.xml format is as follows

The following is a quotation fragment:

<trans name="addDoc" table="TDOC_DOCS" method="insert"> 
　　　　 <primarykey name="docid" />　　　　 
　　　　 <set> 
　　　　　　 <property name="createtime" type="timestamp"/> 
　　　　　　 <property name="creatorid" type="long"/> 
　　　　　　 <property name="doctypeid" type="int"/> 
　　　　　　 <property name="docstatusid" type="int"/>　　　　　　 
　　　　 </set> 
　　 </trans>

The rest of the XML can be customized according to such rules

5. Technical implementation

RouterFactory

The following is a quote:

package com.web.router; 
import com.web.platform.Exception.RouterException; 
import java.util.java/util/Hashtable.java.html" target="_blank">Hashtable;

The following is a quotation fragment:

/** 
 * Router产生和清除的类 
 */ 
public class RouterFactory　 
{ 
　 /** 
　　 * Router存储的树front 
　　 */ 
　 private static java/util/Hashtable.java.html" target="_blank">Hashtable QueuePairFront = null; 
　 /** 
　　 * Router存储的树back 
　　 */ 
　 private static java/util/Hashtable.java.html" target="_blank">Hashtable QueuePairBack = null;  
/** 
　　 * Router存储的树 
　　 */ 
　 private static java/util/Hashtable.java.html" target="_blank">Hashtable QueueRouter = null;  
　 /** 
　　 * 返回的XMLRouter 
　　 */ 
　 public static XMLRouter instance = null; 
　 /** 
　　 * Router的定义 
　　 */ 
　 public static RouterDefine routerdefine = null; 
　 /** 
　　 * Router的ID号 
　　 */ 
　 public static long routeIndex = 0; 
　 /** 
　　 * @roseuid 3F169C21027C 
　　 */ 
　 public RouterFactory()　 
　 { 
　 } 
　 /** 
　　 * 初始化Hashtable和Vector 
　　 */ 
　 public static void initFactory() throws java/lang/Exception.java.html" target="_blank">Exception 
　 { 
　　　　　 QueuePairFront = new java/util/Hashtable.java.html" target="_blank">Hashtable();　 
　　　　　 QueuePairBack　= new java/util/Hashtable.java.html" target="_blank">Hashtable();　 
　　　　　 QueueRouter　　　 = new java/util/Hashtable.java.html" target="_blank">Hashtable(); 
　　　　　 initRouteDefine(); 
　 }  
/** 
　　 * 初始化Route的设置 
　　 *　 
　　 */ 
　 private static void initRouteDefine() throws java/lang/Exception.java.html" target="_blank">Exception 
　 { 
　　　　　 if( routerdefine == null ) 
　　　　　　　 routerdefine = new RouterDefine(); 
　　　　　 routerdefine.loadRouterDef(); 
　 } 
　 /** 
　　 * 返回实例 
　　 * @return com.web.router.XMLRouter 
　　 */ 
　 public static XMLRouter getInstance(long index) throws RouterException 
　 { 
　　　　 return (XMLRouter)QueueRouter.get(new java/lang/Long.java.html" target="_blank">Long(index)); 
　 }  
　 /** 
　　 * 产生一个XMLRouter的实例 
　　 * @return com.web.router.XMLRouter 
　　 * @roseuid 3F1618A103BC 
　　 */ 
　 public static XMLRouter popInstance() throws RouterException 
　 { 
　　　　　 routeIndex ++; 
　　　　　 instance = new XMLRouter(routeIndex); 
　　　　　 setDefine( instance ); 
　　　　　 QueueRouter.put(new java/lang/Long.java.html" target="_blank">Long(routeIndex), instance);　　　　　 
　　　　 return instance; 
　 } 
　 /** 
　　 * 清空Hashtable，Vector等 
　　 * @roseuid 3F1618B203C1 
　　 */ 
　 private static void freeResource() throws java/lang/Exception.java.html" target="_blank">Exception 
　 { 
　　　　　 QueuePairFront.clear(); 
　　　　　 QueuePairBack.clear(); 
　　　　　 QueueRouter.clear(); 
　　　　　 QueuePairFront = QueuePairBack = QueueRouter = null; 
} 
　 /** 
　　 * 清除实例 
　　 * @param instanceID 
　　 * @throws Exception 
　　 */ 
　 public　static void removeInstance(XMLRouter instance) throws java/lang/Exception.java.html" target="_blank">Exception　 
　 { 
　　　　　 instance.clear(); 
　　　　 QueueRouter.remove( new java/lang/Long.java.html" target="_blank">Long(instance.getIndex() ) ) ; 
　 } 
/** 
 * Method isNull. 
 * @return boolean 
 */ 
　 public static boolean isNull() 
　 { 
　　　　 ……　 
　　　　　 return false; 
　 } 
}

XMLRouter

The following is a quotation fragment:

package com.web.router;  
import com.web.platform.Exception.RouterException; 
import com.web.common.*; 
import java.util.*; 
import java.lang.reflect.java/lang/reflect/Method.java.html" target="_blank">Method; 
import java.lang.reflect.java/lang/reflect/Constructor.java.html" target="_blank">Constructor;  
/** 
 * @author keli 
 * @version 0.0.1 
 * 平台的关键，路由的类,每个Router将从RouterFactory里读取 
 * Router存储的树front,和back,routeIndex,目的是为了能在路由 
 * 之后可以清除申请的对象。 
 * Router可以实现同步和异步的功能. 
 */ 
public class XMLRouter　 
{ 
　 /** 
　　 * Router存储的树front 
　　 */ 
　　 private static java/util/Hashtable.java.html" target="_blank">Hashtable QueuePairFront = null; 
　 /** 
　　 * Router存储的树back 
　　 */ 
　　 private static java/util/Hashtable.java.html" target="_blank">Hashtable QueuePairBack = null;  
/** 
　　 * 本router的index号码 
　　 */ 
　　 private long routeIndex = 0;  
/** 
　　 * router的设置 
　　 */ 
　　 private RouterDefine　define = null;　　 
/** 
　　 * 用于判断是路由的起回点 
　　 */ 
　　 private java/lang/String.java.html" target="_blank">String　action = ""; 
　　 /** 
　　 *此变量只是用于在routeto方法中申请新的class　 
　　 */ 
　　 private java/lang/String.java.html" target="_blank">String　classname = ""; 
　　　 /** 
　　 */ 
　　　 public XMLRouter(long index)　 
　　　 { 
　　　　 routeIndex = index; 
　　　 } 
　　　 /** 
　　 * 路由 
　　 * @throws Exception 
　　 * @roseuid 3F1616BD0186 
　　 */ 
　　　 public void routing(Env env) throws RouterException, java/lang/Exception.java.html" target="_blank">Exception　 
　　　 { 
　　　　　 /*如果为起点*/ 
　　　　 if( action.equalsIgnoreCase( RouterConstant.CFG_FUNC_ROUTETO　) )　　 
　　　　 { 
　　　　　　 …… 
　　　　 } 
　　　　 /*如果为返回点*/ 
　　　　 else if( action.equalsIgnoreCase( RouterConstant.CFG_FUNC_ROUTEBACK　) ) 
　　　　 { 
　　　　 …… 
　　　　 } 
　　　　 /*否则为错误*/ 
　　　　 else 
　　　　　　 throw new RouterException("Set Router action error."); 
　　　 } 
　　 /** 
　　 * 读取本Router的id号. 
　　 * @return long 
　　 */ 
　　 public long　getIndex() 
　　 { 
　　　　　 return routeIndex;　　　　 
　　 } 
/** 
　　 * 清除所有对象. 
　　 * @throws RouterException 
　　 */ 
　　 public void clear() throws RouterException 
　　 { 
　　　　 QueuePairFront.remove(new java/lang/Long.java.html" target="_blank">Long(routeIndex)); 
　　　　 QueuePairBack.remove(new java/lang/Long.java.html" target="_blank">Long(routeIndex)); 
　　　　 /*系统回收*/ 
　　　　 java/lang/System.java.html" target="_blank">System.runFinalization();　　　　　 
　　 }  
　　 /** 
　　 * 设置本Router的设置. 
　　 * @param def 
　　 * @throws RouterException 
　　 */ 
　　 public void setDefine(RouterDefine def) throws RouterException 
　　 { 
　　　　 define = def;　　 
　　 }　 
　　 /** 
　　 * 设置action的值 
　　 * @param actionName 
　　 * @throws RouterException 
　　 */ 
　　 public void setAction( java/lang/String.java.html" target="_blank">String actionName ) 
　　 { 
　　　　 action = actionName;　　　　 
　　 } 
}

Service class

The following is a reference fragment:

package com.web.common;  
import com.web.platform.Exception.RouterException;  
/** 
 * Service的父类，abstract　 
 */ 
public abstract class RouteService　 
{ 
　 /** 
　　 */ 
　 public RouteService()　 
　 { 
　 } 
　 /** 
　　 * routeTo方法，是交易的起点。 
　　 * @param env 
　　 * @throws RouterException 
　　 */ 
　 public abstract void routeto(Env env) throws RouterException; 
　 /** 
　　 * routeBack，交易的结束点， 
　　 * @param env 
　　 * @throws RouterException 
　　 */ 
　 public abstract void routeback(Env env) throws RouterException; 
　 /** 
　　 * routeaccept方法，是交易的接收点，也是routeto的接收函数， 
　　 * routeaccept为被动交易对象的主要处理函数 
　　 * @param env 
　　 * @throws RouterException 
　　 */ 
　 public abstract void routeaccept(Env env) throws RouterException;　　 
　 /** 
　　 * routing方法，是Service对外的接口函数 
　　 * @throws RouterException 
　　 */ 
　 public abstract void routing() throws RouterException;

Next, you need to implement all the Services classes, which will not be introduced here.

6. Description

This Router so far It can only implement synchronous transactions and does not support asynchronous transactions for the time being. However, because the Router is designed using the Composite model, asynchronous transactions can also be expanded. I will not do a detailed analysis here.

The above is how to use it. XML realizes the content of the structure of multi-channel access to the website. For more related content, please pay attention to the PHP Chinese website (www.php.cn)!