使用 boost::operators 为模板类定义运算符
我有一个如下所示的类：
template<int DIGITS, int FRACTIONS>
class decimal{
bool operator<(const int& rhs) const;
template<int RDIGITS, int RFRACTIONS>
bool operator<(const decimal<RDIGITS, RFRACTIONS>& rhs) const;
}
我应该能够为整数添加相关的比较运算符
class decimal : boost::less_than_comparable<fixed_decimal<DIGITS, DECSIZE>, int>{ }
但如何将 less_than_comparable 与其他十进制模板一起使用？ 我希望能够比较十进制<10,5>和例如，十进制<6,4>。
I've got a class which looks like this:
template<int DIGITS, int FRACTIONS>
class decimal{
bool operator<(const int& rhs) const;
template<int RDIGITS, int RFRACTIONS>
bool operator<(const decimal<RDIGITS, RFRACTIONS>& rhs) const;
}
I should be able to add related comparison operators for ints with
class decimal : boost::less_than_comparable<fixed_decimal<DIGITS, DECSIZE>, int>{ }
But how can I use less_than_comparable with other decimal templates?
I'd like to be able to compare decimal<10,5> with decimal<6,4>, for example.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
我认为由于 C++ 编译器扩展模板的方式，这是不可能的。您必须为每个
I think this is not possible due to the way C++ compilers expand templates. You'd have to define operators for each
<DIGITS, DECSIZE>-combination. In your place, I'd take a dynamic approach and make those two parameters member variables of your class instead of template parameters.对于返回固定类型的运算符，例如
bool，您可以手动实现运算符：（当然，有很多更聪明的方法来进行比较。）
但是一旦您涉足以下领域更复杂的算术表达式，例如运算符 +，这不再那么容易完成，因为返回类型取决于（目前未知）模板参数。
您可以在C++11中执行如下操作：
...并调用：
如果您不能使用C++11，那么您需要使用模板元函数而不是普通函数/重载运算符。
For operators that return a fixed type such as
bool, you can implement the operators manually:(Of course, there are much cleverer ways to do the comparison.)
But as soon as you foray into the realms of more complex arithmetic expressions such as
operator +this can no longer be done as easily, since the return type depends on (as of yet unknown) template arguments.You can do it in C++11 as follows:
… and to invoke:
If you cannot use C++11, then you need to use template metafunctions instead of normal functions / overloaded operators.