继承实现单例模式的探索（二）

前言

本篇文章继续探索通过继承实现单例模式的可行方案，这次的方案将采用反射机制隐式创建派生类实例，示例代码为C#。

代码

v1.0

using System.Reflection;

/// <summary>
/// 单例模式基类
/// </summary>
/// <typeparam name="T">单例类型</typeparam>
public abstract class Singleton<T>
where T : class
{
    public static T instance => _instance.Value;
    static readonly Lazy<T> _instance = new Lazy<T>(Create);

#pragma warning disable CS8603
    static T Create()
    {
        return Activator.CreateInstance(typeof(T),
        BindingFlags.Public | BindingFlags.NonPublic | BindingFlags.Instance,
        null, null, null) as T;
    }
#pragma warning restore CS8603
}

测试

单例模式基类

/// <summary>
/// 单例模式基类
/// </summary>
/// <typeparam name="T">单例类型</typeparam>
public abstract class SingletonWithTest<T>
where T : class
{
    public static T instance => _instance.Value;
    static readonly Lazy<T> _instance = new Lazy<T>(Create);

#pragma warning disable CS8603
    static T Create()
    {
        return Activator.CreateInstance(typeof(T),
        BindingFlags.Public | BindingFlags.NonPublic | BindingFlags.Instance,
        null, null, null) as T;
    }
#pragma warning restore CS8603
}

派生类

public class TestA : SingletonWithTest<TestA>
{
    public readonly string key;

    TestA() { key = "Default A"; }
}

public class TestB : SingletonWithTest<TestB>
{
    public readonly string key;

    public TestB() { key = "Default B"; }
}

public class TestC : SingletonWithTest<TestC>
{
    public readonly string key;

    public TestC(string key) { this.key = key; }
}

public class TestD : SingletonWithTest<TestD>
{
    public readonly string key;

    public TestD() { key = "Default D"; }
    public TestD(string key) { this.key = key; }
}

public class TestE : SingletonWithTest<TestE>
{
    public readonly string key;

    TestE() { key = "Default E"; }
    public TestE(string key) { this.key = key; }
}

public class TestF : SingletonWithTest<TestF>
{
    public readonly string key = "Default F";
}

测试代码

// ********************************* 线程安全测试：通过 *********************************

Thread t1, t2, t3, t4, t5, t6;

// 打印同一单例的 HashCode 测试：通过
// t1 = new Thread(() => Console.WriteLine("Thread1:" + TestA.instance.GetHashCode()));
// t2 = new Thread(() => Console.WriteLine("Thread2:" + TestA.instance.GetHashCode()));
// t3 = new Thread(() => Console.WriteLine("Thread3:" + TestA.instance.GetHashCode()));
// t4 = new Thread(() => Console.WriteLine("Thread4:" + TestA.instance.GetHashCode()));
// t5 = new Thread(() => Console.WriteLine("Thread5:" + TestA.instance.GetHashCode()));
// t6 = new Thread(() => Console.WriteLine("Thread6:" + TestA.instance.GetHashCode()));

// 同时使用不同单例的测试：通过
t1 = new Thread(() => Console.WriteLine("Thread1:" + TestA.instance.GetHashCode()));
t2 = new Thread(() => Console.WriteLine("Thread2:" + TestB.instance.GetHashCode()));
t3 = new Thread(() => Console.WriteLine("Thread3:" + TestC.instance.GetHashCode())); // 没有无参构造函数，触发异常
t4 = new Thread(() => Console.WriteLine("Thread4:" + TestD.instance.GetHashCode()));
t5 = new Thread(() => Console.WriteLine("Thread5:" + TestE.instance.GetHashCode()));
t6 = new Thread(() => Console.WriteLine("Thread6:" + TestF.instance.GetHashCode()));

t1.Start();
t2.Start();
t3.Start();
t4.Start();
t5.Start();
t6.Start();

t1.Join();
t2.Join();
t3.Join();
t4.Join();
t5.Join();
t6.Join();

优缺点分析

优点	1.继承实现单例模式； 2.按需加载，延迟初始化； 3.线程安全； 4.可以通过单例基类规范统一标准； 5.派生类的无参构造函数用于初始化； 6.将派生类实例创建权限交由派生类本身决定，通常，为遵循单例模式的原则应向外部关闭通过new关键字显式调用构造函数的权限，但这对派生类而言并非硬性要求；
缺点	1.反射开销； 2.派生类必须具备无参构造函数，且无参构造函数需要向单例基类提供可进行反射访问的权限；

版本改进

......

系列文章

继承实现单例模式的探索（一）

如果这篇文章对你有帮助，请给作者点个赞吧！

原文地址：https://blog.csdn.net/hgf1037882434/article/details/142613636

免责声明：本站文章内容转载自网络资源，如本站内容侵犯了原著者的合法权益，可联系本站删除。更多内容请关注自学内容网（zxcms.com）！