通过

在属性和方法之间选择

  • 项目

更新：2007 年 11 月

通常，方法代表操作而属性代表数据。属性应像字段一样使用，这意味着属性不应进行复杂的计算，也不应产生副作用。在不违反下列准则的情况下，应考虑使用属性而不是方法，因为属性对于经验较少的开发人员更易于使用。

如果成员表示类型的逻辑属性 (Attribute)，请考虑使用属性 (Property)。

例如，BorderStyle 是一个属性 (Property)，因为边框样式是 ListView 的属性 (Attribute)。

如果属性值存储在进程内存中并且该属性只是用于提供对值的访问，则要使用属性而不是方法。

下面的代码示例阐释了这一准则。EmployeeRecord 类定义对私有字段提供访问的两个属性。在本主题的末尾显示了完整示例。

Public Class EmployeeRecord

    Private employeeIdValue as Integer
    Private departmentValue as Integer

    Public Sub New()
    End Sub

    Public Sub New (id as Integer, departmentId as Integer)
        EmployeeId = id
        Department = departmentId
    End Sub

    Public Property Department as Integer
        Get 
            Return departmentValue
        End Get
        Set 
            departmentValue = value
        End Set
    End Property

    Public Property EmployeeId as Integer
        Get 
            Return employeeIdValue
        End Get
        Set 
            employeeIdValue = value
        End Set
    End Property
    Public Function Clone() as EmployeeRecord
        Return new EmployeeRecord(employeeIdValue, departmentValue)
    End Function
End Class

public class EmployeeRecord
{
    private int employeeId;
    private int department;
    public EmployeeRecord()
    {
    }
    public  EmployeeRecord (int id, int departmentId)
    {
        EmployeeId = id;
        Department = departmentId;
    }
    public int Department
    {
        get {return department;}
        set {department = value;}
    }
    public int EmployeeId
    {
        get {return employeeId;}
        set {employeeId = value;}
    }
    public EmployeeRecord Clone()
    {
        return new EmployeeRecord(employeeId, department);
    }
}

在下列情况下要使用方法而不是属性。

  • 操作比字段集慢数个数量级。即使考虑提供异步版本的操作来避免阻止线程，该操作也很可能因开销太大而不能使用属性。特别是，访问网络或文件系统（一次性初始化除外）的操作最可能是方法，而不是属性。

  • 操作是转换，如 Object.ToString method。

  • 操作在每次调用时都返回不同的结果，即使参数不发生更改也是如此。例如，NewGuid 方法在每次调用时都返回不同的值。

  • 操作具有很大的显而易见的副作用。注意，一般不将填充内部缓存视为是显而易见的副作用。

  • 操作返回内部状态的副本（这不包括在堆栈上返回的值类型对象的副本）。

  • 操作返回一个数组。

如果操作返回一个数组，应使用方法，原因是：要保留内部数组，必须返回数组的深层副本而不是对属性所使用的数组的引用。这一事实加之开发人员将属性视同字段一样使用的事实，可能会导致代码效率十分低下。在下面的代码示例中阐释了这一点，该示例使用属性返回一个数组。在本主题的末尾显示了完整示例。

Public Class EmployeeData

    Dim data as EmployeeRecord()
    Public Sub New(data as EmployeeRecord())
        Me.data = data
    End Sub
    Public ReadOnly Property Employees as EmployeeRecord()
        Get
            Dim newData as EmployeeRecord() = CopyEmployeeRecords()
            Return newData
        End Get
    End Property

    Private Function CopyEmployeeRecords() as EmployeeRecord()
        Dim newData(UBound(data)) as EmployeeRecord
        For i as Integer = 0 To UBound(data)
            newData(i) = data(i).Clone()
        Next i
        Console.WriteLine ("EmployeeData: cloned employee data.")
        Return newData
    End Function
End Class

public class EmployeeData
{
    EmployeeRecord[] data;
    public EmployeeData(EmployeeRecord[] data)
    {
        this.data = data;
    }
    public EmployeeRecord[] Employees
    {
        get 
        {
            EmployeeRecord[] newData = CopyEmployeeRecords();
            return newData;
        }
    }
    EmployeeRecord[] CopyEmployeeRecords()
    {
        EmployeeRecord[] newData = new EmployeeRecord[data.Length];
        for(int i = 0; i< data.Length; i++)
        {
            newData[i] = data[i].Clone();
        }
        Console.WriteLine ("EmployeeData: cloned employee data.");
        return newData;
    }
}

使用此类的开发人员假定属性不会比字段访问的开销大，因此根据这一假定编写了应用程序代码，如下面的代码示例所示。

Public Class RecordChecker
    Public Shared Function  FindEmployees( _
         dataSource as EmployeeData, _
         department as Integer) as Collection(Of Integer)

        Dim storage as Collection(Of Integer) = new Collection(Of Integer)()
        Console.WriteLine("Record checker: beginning search.")
        For i as Integer = 0 To UBound(dataSource.Employees)
            If dataSource.Employees(i).Department = department
                Console.WriteLine("Record checker: found match at {0}.", i)
                storage.Add(dataSource.Employees(i).EmployeeId)
                Console.WriteLine("Record checker: stored match at {0}.", i)
            Else 
                Console.WriteLine("Record checker: no match at {0}.", i)
            End If
        Next i
        Return storage
    End Function
End Class

public class RecordChecker
{
    public static Collection<int> FindEmployees(EmployeeData dataSource, 
             int department)
    {
        Collection<int> storage = new Collection<int>();
        Console.WriteLine("Record checker: beginning search.");
        for (int i = 0; i < dataSource.Employees.Length; i++)
        {
            if (dataSource.Employees[i].Department == department)
            {
                Console.WriteLine("Record checker: found match at {0}.", i);
                storage.Add(dataSource.Employees[i].EmployeeId);
                Console.WriteLine("Record checker: stored match at {0}.", i);
            }
            else 
            {
                Console.WriteLine("Record checker: no match at {0}.", i);
            }
        }
        return storage;
    }
}

注意，在每次循环迭代中都会访问 Employees 属性，在部门匹配时也会访问该属性。每次访问该属性时，就会创建一个 employees 数组的副本，该副本在短暂的使用过后就需要进行垃圾回收。通过将 Employees 实现为一个方法，可以向开发人员表明：与访问字段相比，该操作在计算上的开销较大。开发人员更愿意调用一次方法，并将方法调用的结果进行缓存以便进行处理。

示例

下面的代码示例演示一个假定属性访问在计算上的开销不大的完整应用程序。EmployeeData 类不适当地定义了一个返回数组副本的属性。

Imports System
Imports System.Collections.ObjectModel

Namespace Examples.DesignGuidelines.Properties
    Public Class EmployeeRecord

        Private employeeIdValue as Integer
        Private departmentValue as Integer

        Public Sub New()
        End Sub

        Public Sub New (id as Integer, departmentId as Integer)
            EmployeeId = id
            Department = departmentId
        End Sub

        Public Property Department as Integer
            Get 
                Return departmentValue
            End Get
            Set 
                departmentValue = value
            End Set
        End Property

        Public Property EmployeeId as Integer
            Get 
                Return employeeIdValue
            End Get
            Set 
                employeeIdValue = value
            End Set
        End Property
        Public Function Clone() as EmployeeRecord
            Return new EmployeeRecord(employeeIdValue, departmentValue)
        End Function
    End Class

Public Class EmployeeData

    Dim data as EmployeeRecord()
    Public Sub New(data as EmployeeRecord())
        Me.data = data
    End Sub
    Public ReadOnly Property Employees as EmployeeRecord()
        Get
            Dim newData as EmployeeRecord() = CopyEmployeeRecords()
            Return newData
        End Get
    End Property

    Private Function CopyEmployeeRecords() as EmployeeRecord()
        Dim newData(UBound(data)) as EmployeeRecord
        For i as Integer = 0 To UBound(data)
            newData(i) = data(i).Clone()
        Next i
        Console.WriteLine ("EmployeeData: cloned employee data.")
        Return newData
    End Function
End Class

Public Class RecordChecker
    Public Shared Function  FindEmployees( _
         dataSource as EmployeeData, _
         department as Integer) as Collection(Of Integer)

        Dim storage as Collection(Of Integer) = new Collection(Of Integer)()
        Console.WriteLine("Record checker: beginning search.")
        For i as Integer = 0 To UBound(dataSource.Employees)
            If dataSource.Employees(i).Department = department
                Console.WriteLine("Record checker: found match at {0}.", i)
                storage.Add(dataSource.Employees(i).EmployeeId)
                Console.WriteLine("Record checker: stored match at {0}.", i)
            Else 
                Console.WriteLine("Record checker: no match at {0}.", i)
            End If
        Next i
        Return storage
    End Function
End Class
    Public Class Tester
        Public Shared Sub Main()
            Dim records(2) as EmployeeRecord
            Dim r0 as EmployeeRecord = new EmployeeRecord()
            r0.EmployeeId = 1
            r0.Department = 100
            records(0) = r0
            Dim r1 as EmployeeRecord = new EmployeeRecord()
            r1.EmployeeId = 2
            r1.Department = 100
            records(1) = r1
            Dim r2 as EmployeeRecord = new EmployeeRecord()
            r2.EmployeeId = 3
            r2.Department = 101
            records(2) = r2
            Dim empData as EmployeeData = new EmployeeData(records)
            Dim hits as Collection(Of Integer)= _ 
                RecordChecker.FindEmployees(empData, 100)
            For Each i as Integer In hits
                Console.WriteLine("found employee {0}", i)
            Next i
        End Sub
    End Class
End Namespace

using System;
using System.Collections.ObjectModel;
namespace Examples.DesignGuidelines.Properties
{
    public class EmployeeRecord
    {
        private int employeeId;
        private int department;
        public EmployeeRecord()
        {
        }
        public  EmployeeRecord (int id, int departmentId)
        {
            EmployeeId = id;
            Department = departmentId;
        }
        public int Department
        {
            get {return department;}
            set {department = value;}
        }
        public int EmployeeId
        {
            get {return employeeId;}
            set {employeeId = value;}
        }
        public EmployeeRecord Clone()
        {
            return new EmployeeRecord(employeeId, department);
        }
    }

public class EmployeeData
{
    EmployeeRecord[] data;
    public EmployeeData(EmployeeRecord[] data)
    {
        this.data = data;
    }
    public EmployeeRecord[] Employees
    {
        get 
        {
            EmployeeRecord[] newData = CopyEmployeeRecords();
            return newData;
        }
    }
    EmployeeRecord[] CopyEmployeeRecords()
    {
        EmployeeRecord[] newData = new EmployeeRecord[data.Length];
        for(int i = 0; i< data.Length; i++)
        {
            newData[i] = data[i].Clone();
        }
        Console.WriteLine ("EmployeeData: cloned employee data.");
        return newData;
    }
}

public class RecordChecker
{
    public static Collection<int> FindEmployees(EmployeeData dataSource, 
             int department)
    {
        Collection<int> storage = new Collection<int>();
        Console.WriteLine("Record checker: beginning search.");
        for (int i = 0; i < dataSource.Employees.Length; i++)
        {
            if (dataSource.Employees[i].Department == department)
            {
                Console.WriteLine("Record checker: found match at {0}.", i);
                storage.Add(dataSource.Employees[i].EmployeeId);
                Console.WriteLine("Record checker: stored match at {0}.", i);
            }
            else 
            {
                Console.WriteLine("Record checker: no match at {0}.", i);
            }
        }
        return storage;
    }
}
    public class Tester
    {
        public static void Main()
        {
            EmployeeRecord[] records  = new EmployeeRecord[3];
            EmployeeRecord r0  = new EmployeeRecord();
            r0.EmployeeId = 1;
            r0.Department = 100;
            records[0] = r0;
            EmployeeRecord r1  = new EmployeeRecord();
            r1.EmployeeId = 2;
            r1.Department = 100;
            records[1] = r1;
            EmployeeRecord r2  = new EmployeeRecord();
            r2.EmployeeId = 3;
            r2.Department = 101;
            records[2] = r2;
            EmployeeData empData = new EmployeeData(records);
            Collection<int> hits = RecordChecker.FindEmployees(empData, 100);
            foreach (int i in hits)
            {
                Console.WriteLine("found employee {0}", i);
            }
        }
    }
}

部分版权所有 2005 Microsoft Corporation。保留所有权利。

部分版权所有 Addison-Wesley Corporation。保留所有权利。

有关设计指南的更多信息，请参见 Krzysztof Cwalina 和 Brad Abrams 编著、Addison-Wesley 于 2005 年出版的“Framework Design Guidelines: Conventions, Idioms, and Patterns for Reusable .NET Libraries”（《框架设计指南：可重用 .NET 库的约定、术语和模式》）。

请参见

概念

属性设计

其他资源

成员设计准则

类库开发的设计准则