Originally posted by: Calvin Sun Hainan

I have an error message in my OPL script "Scripting runtime error: cannot convert to an integer, "2777734350"."

This is strange as OPL script should have no problem to handle a big integer value up to 9e+15 based on my previous understanding.

The only possible cause I can think about is the data type "var" not allow values as big as 2777734350.

In my script, I have some calculation like this.

int x = 2780000000;

execute {

             var a;

             ......

              x -= a;

}

It looks like OPL script tries to convert the variable "a" from "var" to "int" to do the calculation for "x". However, that fails for some reason.

Hi,

the integer values you can use in scripting are limited to pow(2,31)-1

So for example

int x=ftoi(pow(2,31)-1);
int x2=x+1000;
 

execute {
x=x-1;   // works fine
x2=x2-1; // does not work
         
           

}

regards

Originally posted by: Calvin Sun Hainan

Thanks. I will change my variable type to float. That should fix the problem.

Originally posted by: CPlex_NooBie

Hi Alex,

What should I do if I have to use "ftoi" with the execute block?
Is there any other function for it?

Hi,

you may use "Opl." in execute blocks:

float f=12000.0;
int x=ftoi(f);

int x2=0;
execute
{
x2=Opl.ftoi(f);
}
 
 assert x==x2;

regards

Originally posted by: CPlex_NooBie

Thanks, Alex!

It worked! 

Originally posted by: Rajasekhar Kadambur

I am using lamda values for solving multi-objective problem of order 43!(factorial) and the results is showing this error. will you explain this how to overcome this error.

Warning:  Bound infeasibility column 'id32'.
Warning:  Bound infeasibility column 'id32'.
Warning:  -Infinity upper bound ignored.