How to check for Nan ??

kir

HI ,

I am struck with this small problem.Would some one help me to find way out ??

Am performing small calculation where I am not sure whether it will return the valid result or not .
For some iteration it returns the NaN (Not a Number). But what i want is to skip such iteration.
ie. if my result NaN for the iteration 5 then I would like to skip that iteration and move to the next .

here is sample.
float result;
for(int i =0; to say some 100)
 { 
    result = some operation;
    if (result==NaN)
    { 
      result= 0;
      }
  sum = sum + result;

}

benja

Float.isNaN(result)

GoToLoop

Fastest way is if (result != result) continue;. ~O)
It's also more cross-mode compatible w/ "JS Mode" than Float.isNaN(). L-)

kir

@benja : thank u @GoToLoop : Thank U

GoToLoop

An alt. version which relies on the ternary conditional operator ?:: :D
https://Processing.org/reference/conditional.html

// forum.Processing.org/two/discussion/13000/how-to-check-for-nan
// GoToLoop (2015-Oct-14)

final float[] VALS = {
  PI, TAU, Float.NaN, sqrt(3.0), EPSILON
};

float r = 0;

for (float f : VALS)  r += f == f? f : 0.0;
//for (float f : VALS)  r += f;

println(r); // 11.156929
exit();

P.S.: Uncomment line for (float f : VALS) r += f; in order to see the NaN result. >-)