枚举和初始化类 [英] Enums and initialising classes

问题描述

我经常遇到这个令人讨厌的怪癖.

I frequently come across this annoying quirk.

说我有类似的东西

enum class Thing {Ex1, Ex2, Ex3}

我有一个Thing实例，并且想根据它是什么来创建某个类，并且最终看起来像这样

I have an instance of a Thing and want to create a certain class depending on what it is, and it ends up looking like this

switch(_thing){
    case Thing::Ex1:
        ptr = new Class1();
        break;

     case Thing::Ex2:
         ptr = new Class2();
         break;

     case Thing::Ex3:
         ptr = new Class3();
         break; 
}

以此类推.用10到20个这样的语句，它看起来非常糟糕.这是不可避免的，还是有什么方法可以改善这一点?

And so on. With 10-20 such statements, it just looks very bad. Is this an inevitability, or is there some way to improve this?

推荐答案

不幸的是，C ++仍然没有反射.如果代码中有多个位置必须使用相同的值到类型的映射，则可以考虑使用久经考验的X宏:

Unfortunately, C++ still doesn't have reflection. You can consider the tried-and-true X macro, if you have multiple places in your code where you'll have to employ the same mapping of values to types:

#define CASES \
    X(Thing::Ex1, Class1) \
    X(Thing::Ex2, Class2) \
    X(Thing::Ex3, Class3) \
    // ...
// ...
switch (_thing) {
#define X(v, T) \
    case (v): \
        ptr = new (T)(); \
        break;
    CASES
#undef X
}
// ... do other things with CASES and a different X macro
// ...
#undef CASES

这篇关于枚举和初始化类的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！