"MyClass mc = MyClass ()" or "MyClass mc"?

Question

What's the difference between

MyClass mc = MyClass(); 

and

 MyClass mc; 

in c ++?

Answer 1

The first calls the copy constructor, with a temporary object as a parameter - MyClass() creates a temporary one.

The second calls the default constructor.

In fact, they are in most cases optimized for the same code, but what is the semantic difference.

---

As Negal noted, the case is slightly different from POD types; when "MyClass" is POD, the second fragment will not initialize the mc value, while the first will.

Answer 2

The first is copy initialization, and the second is the default initialization.

For example, the following code will not compile:

 class MyC { public: MyC(){} private: MyC(const MyC&) {} }; int main() { MyC myc = MyC(); return 0; } 

Answer 3

Custom copy constructor and default constructor.

Answer 4

First create a temp-object via c-tor without arguments , and then call copy-ctor on the object (do not consider any optimizations). The second call to c-tor without arguments , without copying. When optimizing compilers, both cases are equal.

Differences in fundamental types, therefore

 // initialized to something int c; // initialized to int() that is 0 by standard. int c = int(); 

Answer 5

no difference. default call ctor. sugar syntax) no copy ctor !!!!

 class PPP{ public: PPP(PPP&){ std::cout<<"PPP1"<<std::endl; } PPP(const PPP&){ std::cout<<"PPP2"<<std::endl; } PPP(){ std::cout<<"PPP3"<<std::endl; } }; PPP ppp = PPP(); 

and you will only find PPP in the console.