[aortizb]

Hello, I am trying to rewrite the two snippets below using the vector class. Can anyone guide me in how to do it? For example, can I return a pointer to a vector object? is that the right way to proceed? If someone wants to translate the below snippets would be great. Both snippets do the same thing but I want to see the differences that may arise using vector class with the two approaches below. Thanks.

Snippet 1:

int main()
{
  int size, i;
  double* p;

  size = 100;

  p = new double[size];
  for (i = 0; i < size; i++)
    p[i] = 0.0;

  changeArray(p, size);

  for (i = 0; i < size; i++)
    cout << p[i] << endl;

  delete [] p;

  return 0;
}

Snippet 2:

int main()
{
  int size, i;
  double* p;

  size = 100;

  p = changeArray(size); 
  for (i = 0; i < size; i++)
    cout << p[i] << endl;

  delete [] p;

  return 0;
}

changeArray() is as follows:

For snippet 1:

void changeArray(double* arr, int length)
{
  for (i = 0; i < length; i++)
    arr[i] = (i + 1) * (i + 1);
}

For snippet 2:

double* changeArray(int length)
{
  double* temp;

  temp = new double[length];
  for (i = 0; i < length; i++)
    temp[i] = (i + 1) * (i + 1);

  return temp;
}

[oogabooga]

#include <iostream>
#include <vector>
using namespace std;

typedef vector< double > VecType;

VecType * createVector( int size )
{
  VecType *ret = new VecType( size ); // elements init'd to 0

  for (int i = 0; i < ret->size(); i++)
    (*ret)[i] = (i + 1) * (i + 1);

  return ret;
}

void changeVector( VecType & v )
{
    for( int i = 0; i < v.size(); ++i )
      v[i] = (i + 1) * (i + 1);
}

void showVector( VecType & v )
{
    for( int i = 0; i < v.size(); ++i )
        cout << v[i] << ' ';
    cout << endl;
}

int main()
{
    const int size = 10;
    VecType p( size ); // elements init'd to 0

    showVector( p );
    changeVector( p );
    showVector( p );

    VecType *q = createVector( size );
    showVector( *q );
    delete q;

    return 0;
}

[aortizb]

Thanks for your reply. I have tested another two way to do what you did with vectors above. The first is returning a vector object by value which should be expensive for large vectors. The second is passing a vector object by reference in a function that returns a void value. The third option would be what you coded above (i.e., returning a pointer to a vector object). The second and third option seems to be equally fast. But I am unsure if these are the fastest way I can work. Do you know if there is yet a faster way to do it? My concern is because I am working with large scale problems (i.e., large vectors).

Thanks.