Error with 'ostream' and 'istream' in C++ trying to overloading

0 votes
asked Jun 25, 2017 by aguasal

When modifying a TAD withtemplates to make it generic (add templates):

For input / output data I have this code:

std::ostream& operator<<(std::ostream& os, Matrix & m)
{ os << m.rows() << " " << m.columns() << std::endl; os << std::setprecision(4) << std::fixed; for(int i=1; i <= m.rows(); i++) { for(int j=1; j <= m.columns(); j++) { os << m.value(i,j) << " "; } os << std::endl; } return os;
}
std::istream& operator>>(std::istream& is, Matrix& m)
{ int rows, columns; float v; is >> rows >> columns; for (int i=1; i<=rows; i++) { for (int j=1; j<=columns; j++) { is >> v; m.assign(i,j,v); } } return is;
}

The definition (summary):

#ifndef MATRIX_HPP
#define MATRIX_HPP
#include <stdexcept>
#include <iostream>
#include <iomanip>
template<typename E, int R, int C>
class Matriz
{
public: Matrix();
private: E elements_[R][C];
};
#include "matrix.cpp"
#include "matrix_io.cpp"
#endif // MATRIX_HPP

The implementation (summary):

#include <limits>
#include <cmath>
template<typename E, int R, int C>
Matrix<E,R,C>::Matrix()
{ for(int i=0; i<R; i++) { for(int j=0; j<C; j++) { elements_[i][j] = 0; } }
}

Here is the error:

#include "matrix.hpp"
#define Element float
#define MatrixP Matrix<Element, 3, 3>
void tryBuildMatrix()
{ MatrixP m; std::cout << m;
}

This error:

error: cannot bind 'std::ostream {aka std::basic_ostream}' lvalue to 'std::basic_ostream&&' std::cout << m;

Any idea why it occurs?

PS: if I remove the template the code is perfect.

1 Answer

0 votes
answered Nov 29, 2018 by joseph-willcoxson
template<typename E, int R, int C>
std::ostream& operator<<(std::ostream& os, Matrix<E, R, C> & m)
{
...
}
Welcome to Q&A, where you can ask questions and receive answers from other members of the community.
...