EJB Design Patterns2(0分)

Design Pattern II
To avoid the disadvantages associated with Design Pattern I, we introduce the
concept of a value object that encapsulates all the data values associated with an
Entity Bean. The value object, in FoliQuest? terminology, has a “Struct”
appended to it because of their similarity to CORBA Structs.
Value Object code snippet for Company
public class CompanyStruct implements
java.io.Serializable {
public Integer comId;
//Primary Key
public String comName;
public String comDescription;
public java.sql.Timestamp mutationDate;
}
Value Object code snippet for Employee
public class EmployeeStruct implements
java.io.Serializable {
public Integer empId;
//Primary Key
public Integer comId;
//Foreign Key
public String empFirstName;
public String empLastName;
public java.sql.Timestamp mutationDate;
}
Now, the Company and Employee Entity Bean can accept the above Struct as a
parameter for the ejbCreate(). Since the Struct also encapsulates all the
data associated with an Entity Bean, only one getData() and a setData()
method is required to get and set all the fields in the Entity Bean.
Code snippet for an Entity Bean’s create()
public Integer ejbCreate(CompanyStruct struct) throws
CreateException {
this.comId = struct.comId;
this.comName = struct.comName;
this.comDescription = struct.comDescription;
this.mutationDate = struct.mutationDate;
return null;
}
Code snippet for an Entity Bean’s getData()
public CompanyStruct getData() {
CompanyStruct result = new CompanyStruct();
result.comId = this.comId;
result.comName = this.comName;
result.comDescription = this.comDescription;
result.mutationDate = this.mutationDate;
return result;
}
Code snippet for an Entity Bean’s setData()
public void setData(CompanyStruct struct) {
this.comName = struct.comName;
this.comDescription = struct.comDescription;
this.mutationDate = struct.mutationDate;;
}
Unlike Design Pattern I, where individual get() and set() operations had to be
performed for getting or setting a particular field, Design Pattern II eliminates these
with just a single RPC. Also, now only one transaction is associated with an RPC that
manipulates all the data in one shot. Thus most of the disadvantages associated with
Design Pattern I have been eliminated, except for the modeling of the relationship
between the entity beans.
Though the setData() method sets all the values, the Borland? AppServer?
provides a “tuned” update feature (turned on by default) whereby only the fields that
are actually modified are written to the database. If no fields were modified, then
the ejbStore() is skipped. Refer to the Borland? Programmer’s guide for EJBs
for further details regarding tuned updates.
Also, there is code repetition in the Entity Bean and the Struct, i.e. the same fields
are declared in both classes. This means that any change to the database table
structure will mean a change to both the Entity Bean and the Struct. Code
maintenance becomes difficult as the Struct and the Entity Bean fields have to be
synchronized.
A small improvement can be made to the size of the code in Design Pattern II by
having the ejbCreate() call the setData() method, thus removing some
redundant code.
Code snippet for an Entity Bean’s create()
public Integer ejbCreate(CompanyStruct struct) throws
CreateException {
this.comId = struct.comId;
//set the primary key
setData(struct);//this removes some redundant code
return null;
}

应杜宝的要求，贴出来，望大家指正。

设计模式2
为了避免设计模式1的缺点，我们介绍一下封装
entity bean值域的value objec的概念。value object，
用某些语言的术语来说，就是一个结构类型，因为他们
和corba的结构类型非常类似。

Value Object code snippet for Company
public class CompanyStruct implements
java.io.Serializable {
public Integer comId;
//Primary Key
public String comName;
public String comDescription;
public java.sql.Timestamp mutationDate;
}
Value Object code snippet for Employee
public class EmployeeStruct implements
java.io.Serializable {
public Integer empId;
//Primary Key
public Integer comId;
//Foreign Key
public String empFirstName;
public String empLastName;
public java.sql.Timestamp mutationDate;
}
现在，公司和雇员的entity bean可以把上面的一个结构类型作为
ejbCreate()的一个参数。由于这个结构封装了entity的所有字段
的值，entity bean只需要一个getdata()和setdata()方法就可以
对所有的字段进行操作。

Code snippet for an Entity Bean’s create()
public Integer ejbCreate(CompanyStruct struct) throws
CreateException {
this.comId = struct.comId;
this.comName = struct.comName;
this.comDescription = struct.comDescription;
this.mutationDate = struct.mutationDate;
return null;
}
Code snippet for an Entity Bean’s getData()
public CompanyStruct getData() {
CompanyStruct result = new CompanyStruct();
result.comId = this.comId;
result.comName = this.comName;
result.comDescription = this.comDescription;
result.mutationDate = this.mutationDate;
return result;
}
Code snippet for an Entity Bean’s setData()
public void setData(CompanyStruct struct) {
this.comName = struct.comName;
this.comDescription = struct.comDescription;
this.mutationDate = struct.mutationDate;;
}

跟设计模式1中使用单独的get()和set()方法去操作特定字段不同，
在设计模式2中，我们避免这种情况而只需要进行一次远程调用就
可以了。现在，只有一个事务通过一次远程调用就操作了所有的数
据。这样，我们就避免了设计模式1的大部分缺点，除了建立bean
之间的关系外。
虽然setdata()方法可以对所有字段赋值，但是，borland appserver
提供了一种智能更新的特性，只有被修改过的字段才会被重新写入数
据库，如果没有字段被修改，那么ejbStore()方法将会被跳过。
borland程序员开发指南（EJB）有更详细的描述。
同样，在entity bean和struct之间存在这重复的代码，比如同
样的字段声明。这意味着任何数据库表结构的修改都会导致
entity beabn和struct的改变，这使得同步entity和struct变得
困难起来。
一个小小的改进可以从一定程度上避免这种情况，
就是在ebCreate()方法中调用setddata()方法，这可以消除一
些冗余的代码。

Code snippet for an Entity Bean’s create()
public Integer ejbCreate(CompanyStruct struct) throws
CreateException {
this.comId = struct.comId;
//set the primary key
setData(struct);//this removes some redundant code
return null;
}

接受答案了.