Thread: Weird problem -- negative doubles not being processed by my code

I'm trying to write a very basic if() conditional statement, which in its ideal/simplest form, should look like this:

Code:

 

double Lq[5] = {0.0}; 
double Lq_sign = 0.0; 

/* call function to assign values to Lq[ ] array */ 

for(j = 0; j < 5; j++){ 
     Lq_sign = 0.0; 

     if(Lq[j] >= 0.005) Lq_sign = 1.0;
     else if(fabs(Lq[j]) < 0.005) Lq_sign = 0.0;    
     else if(Lq[j] <= -0.005) Lq_sign = -1.0; 

}

When I print the resulting Lq_sign values for each iteration, the values of 1.0 and 0.0 are correctly assigned with no problems. However, the assignment of -1.0 never happens (the Lq values that *should* be -1.0 show up as 0.0).

I've done some troubleshooting, to determine why the negative Lq values aren't being recognized. Below are the various troubleshooting statements that I've performed:

Code:

 

...
else if(fabs(Lq[j]) > 0.005) printf("abs val is > 0.005.\n"); /* A */
else if(Lq[j] < 0.0) printf("Lq[%d] is < 0.\n", j);   /* B */ 
else if((fabs(Lq[j]) > 0.005) && (Lq[j] < 0)) Lq_sign = -1.0; /* C */

Statement A above always executes when it should, whether the Lq value in question is + or --. However, statement B *never* executes, although there are plenty of negative Lq values that should trigger this statement. And, of course, statement C never executes.

Does anyone have a suggestion about why my negative Lq[ ] array values aren't being recognized as negative?? Thanks in advance for your input.

Do you know what abs does? (Other than make me look ripped?) It makes a number not negative! Why would you ever reach B?

Code:

if number > 0.0005 then
    ... 
else
if( FABS < 0.0005 )
...

The only thing you have not covered here is an exact match for 0.005. I think the only way you could possibly reach B is if you enter -0.0005. Have you tried that?


Quzah.

Originally Posted by CodeKate

Aha! I hadn't been aware that calling fabs( ) was actually *changing* the value!

What is it you think it actually does? If you give it -1.3, it returns 1.3. But it doesn't actually change what you pass it into something else:

Code:

#include<math.h>
#include<stdio.h>
int main( void )
{
    float f = -1.3;
    if( fabs( f ) > 1.0 )
        printf( "f is still %f\n", f );

    return 0;
}

It doesn't actually change what f is.


Quzah.