tranquil hemlock
imagine we have a class "complex"
question: why do we use "&" in definition of two part operations but not in definition of one part operations, or am i wrong or somen (like we can do both?)?
e.g. :
class complex {
double real, imaginary;
public:
complex& operator+=(complex x) { a calculation }
complex operator+(complex x, complex y) { another calculation }
}
radiant echo
so you can something like this
int a = 3;
(a += 4) += 5;
in other words, you can chain them
nothing requires you to return a reference, you could return void but that would make it impossible to objects of your class like this
radiant echo
since the expression (a += 4) returns a reference to a, you can use another operator on it
likewise
struct something {
something& operator+=(something s);
void operator-=(something s);
}
int main(){
something s1, s2, s3;
(s1 += s2) += s3; // ok, since += returns a reference
(s1 -= s2) -= s3; // error, can't use -= with operands of type 'void' and something
}