深度剖析Duwamish 7.0 (1--数据结构)
2024-07-21 02:25:27
供稿：网友
不好意思啊，让大家久等了，第一次写这种文章，可能没有开心那么专业，废话少说，我们开始：
1．结构概述
duwamish 7.0 结构分为四个逻辑层：
web 层
web 层为客户端提供对应用程序的访问。这一层是作为 duwamish.sln 解决方案文件中的 web 项目实现的。web 层由 asp.net web 窗体和代码隐藏文件组成。web 窗体只是用 html 提供用户操作，而代码隐藏文件实现各种控件的事件处理。
业务外观层
业务外观层为 web 层提供处理帐户、类别浏览和购书的界面。这一层是作为 duwamish.sln 解决方案文件中的 businessfacade 项目实现的。业务外观层用作隔离层，它将用户界面与各种业务功能的实现隔离开来。除了低级系统和支持功能之外，对数据库服务器的所有调用都是通过此程序集进行的。
业务规则层
业务规则层是作为 duwamish.sln 解决方案文件中的 businessrules 项目实现的，它包含各种业务规则和逻辑的实现。业务规则完成如客户帐户和书籍订单的验证这样的任务。
数据访问层
数据访问层为业务规则层提供数据服务。这一层是作为 duwamish.sln 解决方案文件中的 dataaccess 项目实现的。
在这里，对于duwamish 7.0的分布式结构我就不再罗嗦了，msdn写的绝对比我好..
下面就从数据访问层开始解剖,我单独提出customer这一段进行程序讲解：
1.customerdata.cs（数据层）
code:
//----------------------------------------------------------------
// copyright (c) 2000-2001 microsoft corporation
// all rights reserved.
//
// this source code is intended only as a supplement to microsoft
// development tools and/or on-line documentation. see these other
// materials for detailed information regarding microsoft code samples.
//
// this code and information is provided "as is" without warranty
// of any kind, either expressed or implied, including but not
// limited to the implied warranties of merchantability and/or
// fitness for a particular purpose.
//----------------------------------------------------------------
namespace duwamish7.common.data
{
using system;
using system.data;
using system.runtime.serialization;
/// <summary>
/// a custom serializable dataset containing customer information.
/// <remarks>
/// this class is used to define the shape of customerdata.
/// </remarks>
/// <remarks>
/// the serializale constructor allows objects of type customerdata to be remoted.
/// </remarks>
/// </summary>
[serializableattribute]
public class customerdata : dataset
{
//
//customer constants
//
/// <value>the constant used for customers table. </value>
public const string customers_table = "customers";
/// <value>the constant used for email field in the customers table. </value>
public const string email_field = "email";
/// <value>the constant used for name field in the customers table. </value>
public const string name_field = "name";
/// <value>the constant used for address field in the customers table. </value>
public const string address_field = "address";
/// <value>the constant used for country field in the customers table. </value>
public const string country_field = "country";
/// <value>the constant used for password field in the customers table. </value>
public const string password_field = "password";
/// <value>the constant used for phonenumber field in the customers table. </value>
public const string phone_field = "phonenumber";
/// <value>the constant used for fax field in the customers table. </value>
public const string fax_field = "fax";
/// <value>the constant used for pkid field in the customers table. </value>
public const string pkid_field = "pkid";
//
// error messages
//
/// <value>the constant used for row error when 'email not unique' field in customerdata. </value>
public const string email_field_not_unique = "email not unique";
/// <value>the constant used for row error when 'email invalid format' field in customerdata. </value>
public const string email_field_invalid_format = "email invalid format";
/// <value>the constant used for row error when there is an 'invalid field' in customerdata. </value>
public const string invalid_field = "invalid field";
/// <value>the constant used for error when 'invalid fields' exist in customerdata. </value>
public const string invalid_fields = "invalid fields";
/// <summary>
/// constructor to support serialization.
/// <remarks>constructor that supports serialization.</remarks>
/// <param name="info">the serializationinfo object to read from.</param>
/// <param name="context">information on who is calling this method.</param>
/// </summary>
public customerdata(serializationinfo info, streamingcontext context) : base(info, context)
{
}
/// <summary>
/// constructor for customerdata.
/// <remarks>initialize a customerdata instance by building the table schema.</remarks>
/// </summary>
public customerdata()
{
//
// create the tables in the dataset
//
builddatatables();
}
//----------------------------------------------------------------
// sub builddatatables:
// creates the following datatables: customers
//----------------------------------------------------------------
private void builddatatables()
{
//
// create the customers table
//
datatable table = new datatable(customers_table);
datacolumncollection columns = table.columns;
datacolumn column = columns.add(pkid_field, typeof(system.int32));
column.allowdbnull = false;
column.autoincrement = true;
columns.add(email_field, typeof(system.string));
columns.add(password_field, typeof(system.string));
columns.add(name_field, typeof(system.string));
columns.add(address_field, typeof(system.string)).allowdbnull= false;
columns.add(country_field, typeof(system.string)).allowdbnull= false;
columns.add(phone_field, typeof(system.string)).allowdbnull= false;
columns.add(fax_field, typeof(system.string));
this.tables.add(table);
}
} //class customerdata
} //namespace duwamish7.common.data
大家可以看到，这里的customerdata类继承与dataset，详细定义字段名称以及有可能产生的出错信息，
并同时绘制出一张空表格。
ok,这里的代码非常简单，大家一看应该就可以明白。代码的技巧性非常强，以后的数据访问、交换、传递全部通过这里进行。
私底下认为，采用这种方法有利于团队协作，代码编辑也很简单。
但是，微软自己也说过dataset的效率没有datareader高，而dataset的灵活性远远超过于datareader，真搞不懂ms在企业事例中为什么采用dataset，也许只是让我们学习其中的思想吧。
过两天再和大家谈谈另一种方法，采用datareader，效率高点，但是代码太难看，难懂：（
