Two dimensional array read from file as argument - c++

Question

I want to read Graph from file and start function to get max flow but i get some error when i pass this graph as function argument. What am I doing wrong?
int main() {
fstream file;
file.open( "macierz.txt", ios::in );
int n = 7;
int graph[n][n];
int v;
for(int i = 0; i < n; i++){
for(int j = 0 ; j < n; j++){
file >> v;
graph[i][j] = v;
}
}
cout << "Ford-Fulkerson -MATRIX- The maximum possible flow: " << fordFulkersonMatrix(graph, 0, 6) << endl;
function :
int fordFulkersonMatrix(int graph[7][7], int start, int target) {
int u, v;
int rGraph[7][7];
int parent[7];
int max_flow = 0;
for (u = 0; u < 7; u++) {
for (v = 0; v < 7; v++)
{
rGraph[u][v] = graph[u][v];
}
}
Error:
main.cpp|200|error: cannot convert 'int (*)[(((unsigned int)(((int)n) + -0x000000001)) + 1)]' to 'const int (*)[7]' for argument '1' to 'int fordFulkersonMatrix(const int (*)[7], int, int)'|

Answer 1

If you program in Java, for sure you will like the C++ STL interfaces:
// or even better - use Eigen as Peter K mentioned
using MyCustomMatrix = std::vector<std::vector<int> >;
// input by const ref to prevent std::vector copy
int fordFulkersonMatrix(const MyCustomMatrix& input, int start, int target) {
// do whatever you need to do
// you can access your matrix elements like this:
int elem34 = input[3][4]; // only if your matrix if big enough, of course
// you can iterare using C++11 range loops
for (const auto& row : input) {
for(auto elem : row) {
std::cout << "My elem: " << elem << std::endl;
}
}
// and you can access your matrix dim information like this:
int numRows = input.size();
if (numRows > 0) {
int numCols = input[0].size();
}
return 0;
}
Later in the code, you can construct your matrix like that:
MyCustomMatrix mat;
mat.push_back(MyCustomMatrix::value_type()); // add one row
mat[0].push_back(69); // add one elem to the first row
fordFulkersonMatrix(mat, 0, 6);
Expected output:
My elem: 69
This is of course very lazy solution, but maybe you don't need anything more. For some serious linear algebra operations consider using specialized libraries (like already mentioned Eigen, or boost uBLAS).