I am a novice C++ programmer, but I thought I know enough about C++ until today when I came across code like this at work and failed to understand how it actually works.
class Object
{
};
template <
class PropObject,
class PropType,
PropType PropObject::* Prop
>
class PropReader
{
public:
void print(Object& o)
{
PropObject& po = static_cast<PropObject &>(o);
PropType& t = po.*Prop;
cout << t << "\n";
}
};
class Student : public Object
{
public:
int age;
int grade;
};
int _tmain(int argc, _TCHAR* argv[])
{
Student s;
s.age = 10;
s.grade = 5;
PropReader<Student, int, &Student::age> r;
PropReader<Student, int, &Student::grade> r2;
r.print(s);
r2.print(s);
}
I think I kind of understood at a high level. But this particular PropType PropObject::* Prop
in the template declaration bothers me. What does it mean? I am looking for an explanation from C++ experts. I would like to understand it, so that I can use it better. It looks very useful though.
A template is not a class or a function. A template is a “pattern” that the compiler uses to generate a family of classes or functions.
" typename " is a keyword in the C++ programming language used when writing templates. It is used for specifying that a dependent name in a template definition or declaration is a type.
Templates in c++ is defined as a blueprint or formula for creating a generic class or a function. Generic Programming is an approach to programming where generic types are used as parameters in algorithms to work for a variety of data types.In C++, a template is a straightforward yet effective tool.
Function templates are special functions that can operate with generic types. This allows us to create a function template whose functionality can be adapted to more than one type or class without repeating the entire code for each type. In C++ this can be achieved using template parameters.
C++ templates are well-known for taking types as arguments, but they can also be parameterized over other types of data as well. For example, you could templatize a class over an integer, as shown here:
template <typename T, unsigned int N> class Array {
private:
T array[N];
public:
/* ... */
};
Templates can also be parameterized over pointers, as long as the pointer meets certain criteria (for example, it has to evaluate to an address of something that can be determined at compile-time). For example, this is perfectly legal:
template <int* Pointer> class ThisIsLegal {
public:
void doSomething() {
*Pointer = 137;
}
};
In your code, the template is parameterized over a pointer-to-class-member. A pointer-to-class-member is similar to a pointer in that it indirectly refers to some object. However, instead of pointing to an object, instead it points to a field in a class. The idea is that you can dereference a pointer-to-class-member relative to some object to select that field out of the class. Here's a simple example of pointers-to-class-member:
struct MyStruct {
int x, y;
};
int main() {
MyStruct ms;
ms.x = 137;
ms.y = 42;
int MyStruct::* ptr; // Declare a pointer to a class member.
ptr = &MyStruct::x; // Now points to the field 'x'
ms.*ptr = 0; // Set the 'x' field of ms to be zero.
}
Notice that the syntax for declaring a pointer-to-class-member is
Type ContainingClass::* pointerName;
So in the above code, int MyStruct::* ptr
means "a pointer to an int
inside of a MyStruct
class.
In the code that you've posted, the template declaration reads like this:
template <
class PropObject,
class PropType,
PropType PropObject::* Prop
>
class PropReader
Let's see what this means. The first two template argument object whose property is going to be read, and PropType
, the type of that property." The final argument to the template is a pointer-to-class-member named Prop
that points inside a PropObject
at a field of type PropType
. For example, you could instantiate this template with MyStruct
like this:
PropReader<MyStruct, int, &MyStruct::x> myPropReader;
Now, let's see what the rest of the code does. The body of this class template is reprinted here:
void print(Object& o)
{
PropObject& po = static_cast<PropObject &>(o);
PropType& t = po.*Prop;
cout << t << "\n";
}
Some of this can be read pretty easily. The parameter to this function is a reference to an Object
named o
, and the last line prints out some field. These two lines are tricky:
PropObject& po = static_cast<PropObject &>(o);
PropType& t = po.*Prop;
This first line is a typecast that says "try to cast the argument o
to a reference of type PropObject
. The idea, I'm guessing, is that Object
is some base class of a lot of different objects. The parameter to the function is just a plain Object
, and this cast tries to convert it to something of the appropriate type (recall that PropObject
is the template argument saying what the type of the object is). Because this uses static_cast
, if the conversion isn't defined (for example, you tried to instantiate the template over int
or vector<string>
), the code won't compile. Otherwise, the code trusts that the cast is safe, then gets a reference of type PropObject
to what the parameter refers to.
Finally, the last line is
PropType& t = po.*Prop;
This uses the pointer-to-class-member dereference syntax I mentioned earlier to say "select the field pointed at by Prop
(the template argument), then store a reference to it named t
.
So, in short, the template
print
function that given an object tries to print out that field.Whew! That was tricky! Hope this helps!
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With