an oddity with ones complement or TTWIW?

[asmboy]

PCM ver 4.066

the ones complement seems odd - or perhaps its me

for the following simple case

unsigned int8 myvar=7;

myvar=~myvar; // generates COMF assembler as desired

~myvar; //
// and
~~myvar; // generates NEITHER an error NOR the desired COMF opcode
// NO code produced at in the LST file for either form

// meanwhile
myvar~; // and
myvar~~; // both generate an error 'expecting ;'


is it ME or the compiler ???

[Ttelmah]

It wouldn't. You are not putting the result into a variable, so the optimiser removes the lines...
It is a bit like saying:

myval+1;

Again no code will be generated.

Best Wishes

[asmboy]

sorry - i guess i was stuck under the impression that 1's complementing was a unitary operation in the exact same category as ++var or --var, which don't need a reassignment to work - since there is an underlying assembler, single cycle opcode for it .
That and the lack of an error when i try ~~var .

i guess you are maybe saying TTWIW is the real answer, since there is no reason a compiler couldn't be made to recognize ~~ and treat it like ++, -- etc.

[SherpaDoug]

What would you expect ~~var to do? Would it be the same as ~(~var) which would do nothing?

++ and -- have clear meanings, but ~~ is not clear to me.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[PCM programmer]

Vs. 4.066 generates a warning. You need to enable warnings.