Originally posted by: SystemAdmin

Hi all,
I need to create a 2-d array for storing only float calculated constants. I tried to create it in two ways as follows:

1)

IloArray<IloNumArray> uSum0(env, h); 

for (

int i = 0; i < h; i++) 
{ uSum0[i] = IloNumArray(env, l);

This way works fine!!

2)

IloArray<IloArray<IloNum>> uSum(env, h); 

for (

int i = 0; i < h; i++) 
{ uSum[i] = IloArray<IloNum> (env, l); 

for (

int s = 0; s < l; s++) 
{ uSum[i][s] = IloNum(env); 
} 
}

But this way gave me error: error C2440: '<function-style-cast>' : cannot convert from 'IloEnv' to 'IloNum' during compilation.
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

I'm not positive (I work with Java, not C++), but I'm pretty sure that IloNum is just an alias for double, does not have a constructor, and is not tied to an environment. So I think the compiler is looking at "IloNum(env)" as an attempt to cast env to an IloNum.

You might want to look at one or two of the C++ examples that ship with CPLEX to see how IloNum is used in practice. When I was <nightmare>coding in C++</nightmare>, I don't think I ever used IloNum; I just used ordinary doubles.

Paul

Mathematicians are like Frenchmen: whenever you say something to them, they translate it into their own language, and at once it is something entirely different. (Goethe)
#DecisionOptimization
#MathematicalProgramming-General

Originally posted by: SystemAdmin

Paul is right. The statement

IloNum(env)

is the problem. IloNum is a primitive type (in fact it is double). Such types do not have a constructor that takes an IloEnv argument. They have a constructor but the only argument the constructor takes is a number.
So you should rewrite this as

uSum[i][s] = IloNum(0);

or just

uSum[i][s] = 0;

#DecisionOptimization
#MathematicalProgramming-General