Logical Not operator in script PLCs?

[JeffLowe]

I had expected to be able to use the unary negation operator (!) yet I can find no documentation supporting this and it causes an error. How do I do a logical negation in a script plc?

[bradp]

You need to use parenthesis around the condition. open plc 1 if(!(p2)){ p1++; } close

[JeffLowe]

[quote='brad' pid='156' dateline='1241479380'] You need to use parenthesis around the condition. open plc 1 if(!(p2)){ p1++; } close [/quote] This seems to only work within conditionals? Still have problems. Consider setting up a t-flop: state = !state; //fail state = !(state); //fail state != state; //fail state += 1 // works, but at best is "trick" programming to be avoided it also leaves not havint a unary negation in complex logical assignments.

[bradp]

In Power PMAC script we do not have a Type Bool. So this brings up some questions concerning how does ! peroform on Doubles. If we follow the "C" world then we have a perfect example of some of the problems. On an x86 you can do ! with doubles. But in the powerpc architecture it works differently. I do specifically remember 20yrs ago a professor saying don't do that kind of stuff because it won't work on some computers. Here is an example #include int main(void) { double a = 0.0; int b = 0; b = !b; printf("b = %d\r\n",b); b = 34; b = !b; printf("b = %d\r\n",b); a = !a; printf("a = %lf\r\n",a); a =4.2; a = !a; printf("a = %ld\r\n",a); }; x86 Output ---------- root@lt-linuxhenry:/opt/eldk-4.2/debian_gdm/root# ./a.out b = 1 b = 0 a = 1.000000 a = 0 PowerPC Output ----------- root@10.34.9.210:~# ./a.out b = 1 b = 0 a = 1.000000 a = 267339300 This output goes back to the ~ function. In Power PMAC script our unary inversion is done with "~", not "!". It is a bit-by-bit inversion. For a Boolean value (is this what you want?) it changes between 1 and 0. So with: M1->*u.1 M1=~(M1) will invert the logical state of M1. With multi-bit variables, it inverts all bits. So with: M2->*u.8 M2=1 M2=~(M2) the value of M2 ends up as 254. We have "~" documented as a bit-by-bit operator. On reflection, it should be listed as a function, because the argument must be in parentheses.