【C#/.NET】record介绍

什么是record？

使用record

record解构

record原理

结论

什么是record？

record是.NET 5中的一种新特性，可以看作是一种概念上不可变的类。records可以帮助我们在C#中更容易地处理数据，同时提供了重要的功能，如对象相等性、hashcode和解构。

与类不同，records具有值语义。也就是说，当比较两个records的实例时，比较的是这些实例的属性而非引用。这意味着，如果两个records的属性值相同，它们就是相等的。

record也可以简化需要类似于Dto的数据结构容器的定义。

使用record

Person p1 = new("小明", "南山", "[email protected]");
Person p2 = new("小明", "南山", "[email protected]");

Console.WriteLine(p1 == p2);
public record Person(string Name, string Address, string Email);

像定义一个类一样,public class Person,只是将class关键字替换成record关键字。然后属性是用括号来定义。

默认的record声明是class，如果想声明一个struct

public record struct Person(string Name, string Address, string Email);

record是不可变的类型，括号中声明的属性在构造之后不可变更。可以使用==按属性的值进行比较。可以直接作为hash的key以及结构。

record可以像普通类一样扩展可变更的属性和自定义的方法，语法如下

public record Person(string Name, string Address, string Email)
{
    public required string PhoneNumber { get; set; }
    public static IEnumerable<Person> GetAll()
    {
        yield return new Person("张三", "123 Main St", "[email protected]") { PhoneNumber = "123456789"};
        yield return new Person("李四", "456 Elm St", "[email protected]") { PhoneNumber = "123456789" };
        yield return new Person("王二", "789 Oak St", "[email protected]") { PhoneNumber = "123456789" }; ;
    }

    public string GetDisplayName() => $"{Name} ({Email})";
};

record解构

record可以通过解构，将对象解构为元组，方便一次性获取record中的属性值，

Person p1 = new("小明", "南山", "[email protected]") ;

var (name,address,email) = p1 ;

record原理

record的原理是编译器提供支持，上述Person定义反编译结果如下

public class Person : IEquatable<Person>
{
    [CompilerGenerated]
    protected virtual Type EqualityContract
    {
        [CompilerGenerated]
        get
        {
            return typeof(Person);
        }
    }

    public string Name { get; set/*init*/; }

    public string Address { get; set/*init*/; }

    public string Email { get; set/*init*/; }

    public Person(string Name, string Address, string Email)
    {
        this.Name = Name;
        this.Address = Address;
        this.Email = Email;
        base..ctor();
    }

    [CompilerGenerated]
    public override string ToString()
    {
        StringBuilder stringBuilder = new StringBuilder();
        stringBuilder.Append("Person");
        stringBuilder.Append(" { ");
        if (PrintMembers(stringBuilder))
        {
            stringBuilder.Append(' ');
        }
        stringBuilder.Append('}');
        return stringBuilder.ToString();
    }

    [CompilerGenerated]
    protected virtual bool PrintMembers(StringBuilder builder)
    {
        RuntimeHelpers.EnsureSufficientExecutionStack();
        builder.Append("Name = ");
        builder.Append((object?)Name);
        builder.Append(", Address = ");
        builder.Append((object?)Address);
        builder.Append(", Email = ");
        builder.Append((object?)Email);
        return true;
    }

    [CompilerGenerated]
    public static bool operator !=(Person? left, Person? right)
    {
        return !(left == right);
    }

    [CompilerGenerated]
    public static bool operator ==(Person? left, Person? right)
    {
        return (object)left == right || (left?.Equals(right) ?? false);
    }

    [CompilerGenerated]
    public override int GetHashCode()
    {
        return ((EqualityComparer<Type>.Default.GetHashCode(EqualityContract) * -1521134295 + EqualityComparer<string>.Default.GetHashCode(Name)) * -1521134295 + EqualityComparer<string>.Default.GetHashCode(Address)) * -1521134295 + EqualityComparer<string>.Default.GetHashCode(Email);
    }

    [CompilerGenerated]
    public override bool Equals(object? obj)
    {
        return Equals(obj as Person);
    }

    [CompilerGenerated]
    public virtual bool Equals(Person? other)
    {
        return (object)this == other || ((object)other != null && EqualityContract == other.EqualityContract && EqualityComparer<string>.Default.Equals(Name, other.Name) && EqualityComparer<string>.Default.Equals(Address, other.Address) && EqualityComparer<string>.Default.Equals(Email, other.Email));
    }

    [CompilerGenerated]
    protected Person(Person original)
    {
        Name = original.Name;
        Address = original.Address;
        Email = original.Email;
    }

    [CompilerGenerated]
    public void Deconstruct(out string Name, out string Address, out string Email)
    {
        Name = this.Name;
        Address = this.Address;
        Email = this.Email;
    }
}

可以看到，编译器给使用了record关键字的定义生成了对应的属性和构造函数，并且重写了ToString(),GetHashCode,Equals还有一个解构函数和!=和==运算符。其实看到这里就明白了，为什么record可以提供值比较，解构，hash等。

不可变性是因为record的属性是使用了init关键字而不是set，这样子如果对record的对象属性赋值，编译器会报错。

值相等性是重定义了!=和==运算符

hash是因为重写了GetHashCode,Equals

解构是定义了Deconstruct方法

结论

我们介绍了.NET 5中引入的record类型及其优点。但对于许多数据对象的简单情况，如值对象和DTO，推荐使用record类型。虽然record可以定义可变更的属性和添加方法，不过这样子有点违背了record的初衷。

欲东