.NET中的结构和类有什么区别?
当前回答
Struct | Class | |
---|---|---|
Type | Value-type | Reference-type |
Where | On stack / Inline in containing type | On Heap |
Deallocation | Stack unwinds / containing type gets deallocated | Garbage Collected |
Arrays | Inline, elements are the actual instances of the value type | Out of line, elements are just references to instances of the reference type residing on the heap |
Al-Del Cost | Cheap allocation-deallocation | Expensive allocation-deallocation |
Memory usage | Boxed when cast to a reference type or one of the interfaces they implement, Unboxed when cast back to value type (Negative impact because boxes are objects that are allocated on the heap and are garbage-collected) |
No boxing-unboxing |
Assignments | Copy entire data | Copy the reference |
Change to an instance | Does not affect any of its copies | Affect all references pointing to the instance |
Mutability | Should be immutable | Mutable |
Population | In some situations | Majority of types in a framework should be classes |
Lifetime | Short-lived | Long-lived |
Destructor | Cannot have | Can have |
Inheritance | Only from an interface | Full support |
Polymorphism | No | Yes |
Sealed | Yes | When have sealed keyword (C#), or Sealed attribute (F#) |
Constructor | Can not have explicit parameterless constructors | Any constructor |
Null-assignments | When marked with nullable question mark | Yes (When marked with nullable question mark in C# 8+ and F# 5+ 1) |
Abstract | No | When have abstract keyword (C#), or AbstractClass attribute (F#) |
Member Access Modifiers | public , private , internal |
public , protected , internal , protected internal , private protected |
1不鼓励在F#中使用null,请改用Option类型。
其他回答
除了其他答案之外,还有一个基本的区别值得注意,那就是数据如何存储在数组中,因为这会对性能产生重大影响。
对于结构,数组包含结构的实例对于类,数组包含指向内存中其他位置的类实例的指针
因此,内存中的结构数组如下所示
[结构][结构]〔结构〕〔结构〕
而类数组看起来像这样
指针
对于一个类数组,您感兴趣的值不会存储在数组中,而是存储在内存的其他位置。
对于绝大多数应用程序来说,这种差异并不重要,但是,在高性能代码中,这将影响内存中数据的位置,并对CPU缓存的性能产生很大影响。在可以/应该使用结构的情况下使用类将大大增加CPU上的缓存未命中数。
现代CPU做的最慢的事情不是处理数字,而是从内存中提取数据,一级缓存命中速度比从RAM读取数据快很多倍。
下面是一些可以测试的代码。在我的机器上,遍历类数组所需的时间大约是结构数组的3倍。
private struct PerformanceStruct
{
public int i1;
public int i2;
}
private class PerformanceClass
{
public int i1;
public int i2;
}
private static void DoTest()
{
var structArray = new PerformanceStruct[100000000];
var classArray = new PerformanceClass[structArray.Length];
for (var i = 0; i < structArray.Length; i++)
{
structArray[i] = new PerformanceStruct();
classArray[i] = new PerformanceClass();
}
long total = 0;
var sw = new Stopwatch();
sw.Start();
for (var loops = 0; loops < 100; loops++)
for (var i = 0; i < structArray.Length; i++)
{
total += structArray[i].i1 + structArray[i].i2;
}
sw.Stop();
Console.WriteLine($"Struct Time: {sw.ElapsedMilliseconds}");
sw = new Stopwatch();
sw.Start();
for (var loops = 0; loops < 100; loops++)
for (var i = 0; i < classArray.Length; i++)
{
total += classArray[i].i1 + classArray[i].i2;
}
Console.WriteLine($"Class Time: {sw.ElapsedMilliseconds}");
}
除了其他答案中描述的所有差异外:
结构不能具有显式无参数构造函数,而类可以结构不能有析构函数,而类可以结构不能从其他结构或类继承,而类可以从其他类继承。(结构和类都可以从接口实现。)
如果您正在观看一段视频,解释所有的差异,您可以查看第29部分-C#教程-C#中类和结构之间的差异。
结构与等级
结构是一种值类型,因此它存储在堆栈上,但类是一种引用类型,存储在堆上。
结构不支持继承和多态,但类同时支持两者。
默认情况下,所有结构成员都是公共的,但类成员在本质上默认是私有的。
由于结构是一种值类型,我们不能将null赋给结构对象,但类的情况并非如此。
Struct | Class | |
---|---|---|
Type | Value-type | Reference-type |
Where | On stack / Inline in containing type | On Heap |
Deallocation | Stack unwinds / containing type gets deallocated | Garbage Collected |
Arrays | Inline, elements are the actual instances of the value type | Out of line, elements are just references to instances of the reference type residing on the heap |
Al-Del Cost | Cheap allocation-deallocation | Expensive allocation-deallocation |
Memory usage | Boxed when cast to a reference type or one of the interfaces they implement, Unboxed when cast back to value type (Negative impact because boxes are objects that are allocated on the heap and are garbage-collected) |
No boxing-unboxing |
Assignments | Copy entire data | Copy the reference |
Change to an instance | Does not affect any of its copies | Affect all references pointing to the instance |
Mutability | Should be immutable | Mutable |
Population | In some situations | Majority of types in a framework should be classes |
Lifetime | Short-lived | Long-lived |
Destructor | Cannot have | Can have |
Inheritance | Only from an interface | Full support |
Polymorphism | No | Yes |
Sealed | Yes | When have sealed keyword (C#), or Sealed attribute (F#) |
Constructor | Can not have explicit parameterless constructors | Any constructor |
Null-assignments | When marked with nullable question mark | Yes (When marked with nullable question mark in C# 8+ and F# 5+ 1) |
Abstract | No | When have abstract keyword (C#), or AbstractClass attribute (F#) |
Member Access Modifiers | public , private , internal |
public , protected , internal , protected internal , private protected |
1不鼓励在F#中使用null,请改用Option类型。
类的实例存储在托管堆上。“包含”实例的所有变量都只是对堆上实例的引用。将对象传递给方法会导致传递引用的副本,而不是对象本身。
结构(从技术上讲,值类型)存储在使用它们的任何地方,很像原始类型。运行时可以随时复制内容,而无需调用自定义的复制构造函数。将值类型传递给方法涉及复制整个值,同样无需调用任何自定义代码。
C++/CLI名称使这种区别更加明显:“ref class”是第一个类,“value class”是第二个类。C#使用的关键字“class”和“struct”只是必须学习的东西。
推荐文章
- 如何为构造函数定制Visual Studio的私有字段生成快捷方式?
- 为什么Visual Studio 2015/2017/2019测试运行器没有发现我的xUnit v2测试
- AppSettings从.config文件中获取值
- 如何检查IEnumerable是否为空或空?
- 没有ListBox。SelectionMode="None",是否有其他方法禁用列表框中的选择?
- 在c#代码中设置WPF文本框的背景颜色
- 如何在iis7应用程序池中设置。net Framework 4.5版本
- 如何分裂()一个分隔字符串到一个列表<字符串>
- 如何指定最小值,但没有使用范围数据注释属性的最大小数?
- 类中的Python装饰器
- 如何在PowerShell中获得本地主机名?
- 为什么在Java和。net中不能修改字符串?
- 'throw'和'throw new Exception()'的区别
- c# int到字节[]
- 如何跟踪log4net问题