Two dimensional arrays error [closed] - c++

Question

Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
This question appears to be off-topic because it lacks sufficient information to diagnose the problem. Describe your problem in more detail or include a minimal example in the question itself.
Closed 9 years ago.
Improve this question
#include<iostream>
using namespace std;
double NUMB_COLS = 3;
void readMatrix(int matr[] [3], int numb_rows, int numb_cols)
{
for (int i = 0; i<numb_rows; i++)
{
for (int j = 0; j < numb_cols; j++)
{
cout <<"[" << i << "] [" <<j <<"] =";
cin >> matr[i][j];
}
}
}
void printMatr(int matr[][3], int numb_rows, int numb_cols)
{
for (int i = 0; i < numb_rows; i++)
{
for (int j = 0; j < numb_cols; j++)
{
cout << "[" << i << "][" << j << "] = " << matr[i][j];
}
cout << endl;
}
}
int main()
{
int matr[5][10];
printMatr(matr, 4, 5);
readMatrix(matr, 4, 5);
return 0;
}
The error is
31 23 C:\Users\Windows\Desktop\programs\arrays2.cpp [Error] cannot convert 'int ()[10]' to 'int ()[3]' for argument '1' to 'void readMatrix(int (*)[3], int, int)'
What to do??

Answer 1

The first error is, that you pass a pointer to matr to a function expecting a different array layout. Remember that all ints in matr are consecutive in memory. Even if you did cast your matr to the expected type before passing it to the function, what was at position matr[0][7] in main() will end up in position matr[2][1] inside readMatrix().
The second error is, that your function accepts a column count, even though it already declared the column count to be 3. Such inconsistencies are an ample source of bugs, and should be avoided at all cost. Unfortunately, C++ forbids you to use dynamically sized array types, so you cannot just change your function declaration to
void readMatrix(int numb_rows, int numb_cols, int matr[numb_rows][numb_cols]) {
as you can in C.
The easiest way to fix your problem is probably to use std::vector<>.

Answer 2

You need to specify the correct subscripts.
void readMatrix(int matr[5][10], int numb_rows, int numb_cols)
a std::vector would be a lot easier in this case.

Answer 3

You defined the array in main as
int matr[5][10];
but in the functions the corresponding parameter is defined as
int matr[][3]
They are different types.
If you want to use in the functions only a part of the array int matr[5][10]; that is only 4 first rows and 5 first columns then you have to define the corresponding parameter of the functions the same way as the original array that is as
int matr[][10];
For example
void printMatr(int matr[][10], int numb_rows, int numb_cols)
Or you have to change code in main and write
int matr[5][3];
printMatr(matr, 5, 3);
readMatrix(matr, 5, 3);
Also take into account that you defined variable
double NUMB_COLS = 3;
but do not use it (thogh it should be defined having integral type).
If you would write
const int NUMB_COLS = 3;
and then use it in array definitions there would not be such an error. For example
#include<iostream>
using namespace std;
const int NUMB_COLS = 3;
void readMatrix(int matr[] [NUMB_COLS], int numb_rows, int numb_cols)
{
for (int i = 0; i<numb_rows; i++)
{
for (int j = 0; j < numb_cols; j++)
{
cout <<"[" << i << "] [" <<j <<"] =";
cin >> matr[i][j];
}
}
}
void printMatr(int matr[][NUMB_COLS], int numb_rows, int numb_cols)
{
for (int i = 0; i < numb_rows; i++)
{
for (int j = 0; j < numb_cols; j++)
{
cout << "[" << i << "][" << j << "] = " << matr[i][j];
}
cout << endl;
}
}
int main()
{
int matr[5][NUMB_COLS];
printMatr(matr, 5, NUMB_COLS);
readMatrix(matr, 5, NUMB_COLS);
return 0;
}
You can assign NUMB_COLS any number including 10 used in your original code. The code I showed will not depend on "magic numbers". It uses only named constant NUMB_COLS