C++

Forum

 
Proper Initialization of member variables

khoder Alshaar (7)
Given a class "MyClass" defined as follows:

1
2
3
4
5
6
7
8
 
class Myclass
{
    public:
    Myclass();
    typedef vector<Myclass> vectorField;
    vector<double> value;
    int N, M;
}

Consider that I have created the following object:
1
2
 
//in main
  Myclass::vectorField Vec;


It seems that I have a problem with initialization. How to properly initialize the class?

The problem appears when I try to access the vector object for any of its member values like the following:
1
2
3
4
5
 
     double value;
     for (int i = 0; i < (10); i++)
     {
        value= Vec[i].M;
     }


The code compiles with GCC, but it terminates upon running. Could you please help me identify the problem and the correct way of initialization?

seeplus (6591)
I'm not sure what you are trying to achieve. Perhaps you could give some more details. However to initialise a struct/class and display some values consider:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
 
#include <iostream>
#include <vector>

class Myclass {
public:
	std::vector<double> value;
	int N {}, M {};
};

using vectorField = std::vector<Myclass>;


int main() {
	const vectorField Vec { {{1, 2, 3}, 2, 3 }, {{ 2, 3, 4 }, 3, 4} };

	for (const auto& v : Vec) {
		for (const auto& va : v.value)
			std::cout << va << ' ';

		std::cout << '\n';
	}
}


which displays:


1 2 3
2 3 4

Peter87 (11235)
This creates an empty vector (containing zero elements).
 
 
Myclass::vectorField Vec;

This creates a vector containing 10 elements.
 
 
Myclass::vectorField Vec(10);

This adds one element to the end of the vector.
 
 
Vec.push_back(Myclass());


When accessing vector elements using [] you need to be sure the vector is big enough. Otherwise the behaviour is undefined.
Last edited on
seeplus (6591)
Also consider:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
 
#include <iostream>
#include <vector>

class Myclass {
public:
	using vectorField = std::vector<Myclass>;

	std::vector<double> value;
	int N {}, M {};
};


int main() {
	Myclass::vectorField Vec { {{1, 2, 3}, 2, 3 }, {{ 2, 3, 4 }, 3, 4} };

	Vec.emplace_back( std::vector<double>{ 5, 6, 7 }, 4, 5 );

	for (size_t i {}; i < Vec.size(); ++i)
		std::cout << Vec[i].M << ' ';

	std::cout << '\n';
}


when using [] with a vector, the specified element number must be >= 0 and < .size().
keskiverto (10402)
Consider that I have created the following object:
1
2
 
//in main
Myclass::vectorField Vec;

That is equivalent to:
1
2
 
//in main
std::vector<Myclass> Vec;

Just like std::vector<int> integers;, the Vec is an empty vector that contains no elements.

The std::vector does have constructors that help you initialize the vector in multiple ways.
Some of them create elements and assign values to those elements, like above:
1
2
3
4
 
Myclass::vectorField Vec {
      {{1, 2, 3}, 2, 3},
      {{2, 3, 4}, 3, 4}
   };

That construction supplies values for two Myclass elements.

The only difference between std::vector<int> and std::vector<Myclass> is the type of element.


If you would create just one Myclass object: Myclass obj ;,
then how would you initialize that?
Topic archived. No new replies allowed.