如何修改给定类以使用 const 运算符
我正在尝试解决我的问题关于在多个级别使用push_back。从评论/答案中可以清楚地看出,我必须:
- 创建一个采用 const 参数的复制运算符
- 将所有运算符修改为 const
但因为这个头文件是给我的,所以有一个运算符我无法将其变为 const。这很简单:
float & operator [] (int i) {
return _item[i];
}
在给定的程序中,该运算符用于获取和设置数据。
我的问题是,因为我需要在头文件中包含此运算符,所以我无法将所有其他运算符转换为 const,这意味着我无法插入复制运算符。
如何将所有运算符都变为 const,同时保留已编写程序的功能?
以下是该类的完整声明:
class Vector3f {
float _item[3];
public:
float & operator [] (int i) {
return _item[i];
}
Vector3f(float x, float y, float z)
{ _item[0] = x ; _item[1] = y ; _item[2] = z; };
Vector3f() {};
Vector3f & operator = ( const Vector3f& obj)
{
_item[0] = obj[0];
_item[1] = obj[1];
_item[2] = obj[2];
return *this;
};
Vector3f & operator += ( const Vector3f & obj)
{
_item[0] += obj[0];
_item[1] += obj[1];
_item[2] += obj[2];
return *this;
};
bool operator ==( const Vector3f & obj) {
bool x = (_item[0] == obj[0]) && (_item[1] == obj[1]) && (_item[2] == obj[2]);
return x;
}
// my copy operator
Vector3f(const Vector3f& obj) {
_item[0] += obj[0];
_item[1] += obj[1];
_item[2] += obj[2];
return this;
}
};
I am trying to solve my question regarding using push_back in more than one level. From the comments/answers it is clear that I have to:
- Create a copy operator which takes a const argument
- Modify all my operators to const
But because this header file is given to me there is an operator what I cannot make into const. It is a simple:
float & operator [] (int i) {
return _item[i];
}
In the given program, this operator is used to get and set data.
My problem is that because I need to have this operator in the header file, I cannot turn all the other operators to const, what means I cannot insert a copy operator.
How can I make all my operators into const, while preserving the functionality of the already written program?
Here is the full declaration of the class:
class Vector3f {
float _item[3];
public:
float & operator [] (int i) {
return _item[i];
}
Vector3f(float x, float y, float z)
{ _item[0] = x ; _item[1] = y ; _item[2] = z; };
Vector3f() {};
Vector3f & operator = ( const Vector3f& obj)
{
_item[0] = obj[0];
_item[1] = obj[1];
_item[2] = obj[2];
return *this;
};
Vector3f & operator += ( const Vector3f & obj)
{
_item[0] += obj[0];
_item[1] += obj[1];
_item[2] += obj[2];
return *this;
};
bool operator ==( const Vector3f & obj) {
bool x = (_item[0] == obj[0]) && (_item[1] == obj[1]) && (_item[2] == obj[2]);
return x;
}
// my copy operator
Vector3f(const Vector3f& obj) {
_item[0] += obj[0];
_item[1] += obj[1];
_item[2] += obj[2];
return this;
}
};
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)
我不太明白你想要做什么,但我注意到这段代码不可能编译。原因是复制是由复制构造函数处理的,而不是操作符。这意味着,像任何构造函数一样,它不会返回任何内容。从构造函数中删除
return
语句,如下所示:至于使运算符
const
,您可以简单地重载它并提供同一方法的两个版本。第一个将是非 const 并将返回一个引用(允许修改),而第二个将是 const 并返回一个副本(理想情况下您应该返回一个>const 引用,但由于float
是原始类型,因此只需按值返回)。I did not really understand what you're trying to do, but I noticed that this code can't possibly compile. The reason is that copy is handled by a copy constructor, not operator. Which means that, like any constructor, it doesn't return anything. Remove the
return
statement from your constructor, like so:As for making your operator
const
, you can simply overload it and offer two versions of the same method. The first one will be non-const
and will return a reference (allowing modifications), while the second will beconst
and return a copy (ideally you should return aconst
reference, but sincefloat
s are primitive types, just return by value).这是很正常的——你创建一个运算符,它既提供常规值的引用,也提供常量值的引用;
此模式适用于原子类型(例如 float、int 等)以及更复杂的结构和类类型。
This is quite normal -- you make an operator which provides both at regular value by reference as well as a const value by reference;
This pattern works for both atomic types such as float,int, etc.., as well as more complex struct and class types.