Originally posted by: JorisK

According to the documentation (Cplex 12.8, java interface), IloLPMatrix.getRows(int start, int num, double[] lb, double[] ub, int[][] ind, double[][] val) returns rows in a sparse matrix representation. Apparently this is not always the case. Take for instance this example:

public static void main(String[] args) throws IloException {
    IloCplex cplex = new IloCplex();

    //PROBLEM DEFINITION
    int numVars=50;
    int numConstraints=6;
    int objCoefficients[]={4, 100, 3, 7, 12, 3, 8, 5, 6, 10, 12, 8, 13, 8, 3, 7, 5, 10, 6, 5, 15, 6, 34, 12, 4, 6, 100, 5, 3, 34, 34, 5, 3, 9, 4, 6, 5, 4, 6, 13, 20, 8, 9, 3, 4, 3, 50, 8, 5, 50};
    int[][] coefficientMatrix={{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 17, 0, 0, 0, 0, 0, 0, 0, 0, 0, 12, 0, 0, 0, 13, 0, 2, 13, 15, 0, 0, 0, 0, 10, 13, 0, 0, 0, 0, 0, 0, 0, 0, 0, 3, 0},
            {0, 0, 0, 0, 0, 13, 0, 0, 9, 0, 0, 1, 0, 9, 0, 11, 17, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 19, 8, 0, 4, 13, 0, 0, 0, 4, 13, 11, 0, 0, 10, 0, 0, 0, 0},
            {0, 0, 17, 15, 0, 0, 0, 11, 0, 10, 7, 0, 0, 0, 0, 1, 0, 8, 0, 0, 0, 0, 0, 0, 15, 0, 0, 20, 16, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8, 19, 0, 1, 0, 0, 1},
            {0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 7, 3, 0, 0, 4, 1, 0, 11, 6, 0, 0, 0, 0, 16, 0, 0, 0, 0, 0, 0, 0, 0, 15, 0, 0, 20, 0, 7, 0, 0, 0, 15, 8, 16, 0, 13, 0, 0},
            {15, 0, 0, 0, 8, 5, 0, 10, 0, 0, 0, 0, 0, 0, 16, 0, 0, 0, 18, 0, 0, 0, 0, 8, 0, 0, 0, 0, 6, 0, 0, 6, 13, 3, 16, 14, 6, 0, 0, 0, 0, 0, 0, 18, 0, 16, 0, 0, 0, 0},
            {17, 0, 17, 0, 0, 20, 12, 0, 9, 0, 0, 12, 0, 0, 0, 3, 0, 0, 0, 10, 0, 18, 2, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 5, 0, 0, 0, 0, 3, 0, 0, 0, 0, 18, 0}};
    int[] rhs={20, 29, 30, 29, 36, 30};

    //CPLEX MODEL

    //Variables
    IloNumVar[] x = new IloNumVar[numVars];
    for (int i = 0; i < numVars; i++)
        x[i] = cplex.boolVar("x_" + i);

    // Objective
    cplex.addMaximize(cplex.scalProd(objCoefficients, x));

    //Constraints
    IloLPMatrix lpMatrix=cplex.LPMatrix();
    for(int k=0; k<numConstraints; k++)
        lpMatrix.addRow(
                cplex.eq(
                        cplex.scalProd(coefficientMatrix[k],x),
                        rhs[k],
                        "C"+k)
        );

    //Result of LPMatrix.getRows method
    int[][] ind=new int[numConstraints][]; //SPARSE table!
    double[][] coefficients=new double[numConstraints][]; //SPARSE table!
    double[] lb=new double[numConstraints];
    double[] ub=new double[numConstraints];

    lpMatrix.getRows(0, numConstraints, lb, ub, ind, coefficients); //populate arrays

    System.out.println("ind[0]: "+ Arrays.toString(ind[0]));
    System.out.println("coefficients[0]: "+ Arrays.toString(coefficients[0]));
}

The output of this little code snippet is:

ind[0]: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49]
coefficients[0]: [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 1.0, 0.0, 0.0, 0.0, 17.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 12.0, 0.0, 0.0, 0.0, 13.0, 0.0, 2.0, 13.0, 15.0, 0.0, 0.0, 0.0, 0.0, 10.0, 13.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 3.0, 0.0]

As opposed to:

ind[0]: [10, 14, 24, 28, 30, 31, 32, 37, 38, 48]
coefficients[0]: [1.0, 17.0, 12.0, 13.0, 2.0, 13.0, 15.0, 10.0, 13.0, 3.0]

As it turns out, a sparse representation is only returned after the LPMatrix has been added to the cplex model. So instead of IloLPMatrix lpMatrix=cplex.LPMatrix();, one should invoke IloLPMatrix lpMatrix=cplex.addLPMatrix(); to get the sparse representation! This is not mentioned in the documentation, and took quite some time to troubleshoot. Please explicitly state this in the documentation (or have the method correctly return a sparse representation at all times).

Thank you for reporting this. I have filed a defect to get this fixed one way or other.