How to: Convert with Constructors
A converting constructor takes a type and uses it to create an object. A converting constructor is called with direct initialization only; casts will not invoke converting constructors.
By default, converting constructors are explicit for CLR types.
Example
// clr_udc_converting_constructors.cpp
// compile with: /clr
public ref struct R {
int m;
char c;
R(int i) : m(i) { }
R(char j) : c(j) { }
};
public value struct V {
R^ ptr;
int m;
V(R^ r) : ptr(r) { }
V(int i) : m(i) { }
};
int main() {
R^ r = gcnew R(5);
System::Console::WriteLine( V(5).m);
System::Console::WriteLine( V(r).ptr);
}
Output
5 R
In this code sample, an implicit static conversion function does the same thing as an explicit conversion constructor.
public value struct V {
int m;
V(int i) : m(i) {}
static operator V(int i) {
V v(i*100);
return v;
}
};
public ref struct R {
int m;
R(int i) : m(i) {}
static operator R^(int i) {
return gcnew R(i*100);
}
};
int main() {
V v(13); // explicit
R^ r = gcnew R(12); // explicit
System::Console::WriteLine(v.m);
System::Console::WriteLine(r->m);
// explicit ctor can't be called here: not ambiguous
v = 5;
r = 20;
System::Console::WriteLine(v.m);
System::Console::WriteLine(r->m);
}
Output
13 12 500 2000