if bug?

[Marcoos]

These are two versions of the same program, the first does not work,
the second compiles correctly
I something wrong or is this a bug?
Tanks, Marco

if (test1<> 0x55)|(test2<> 0xAA) THEN
    Goto SetMeasure
END IF

if (test1<> 0x55) THEN
    Goto SetMeasure
END IF
if (test2<> 0xAA) THEN
    Goto SetMeasure
END IF

[Anobium]

Try the following. Please post the results.

if ((test1<> 0x55) = 1 )| ((test2<> 0xAA) = 1 )THEN
Goto SetMeasure
END IF

[Anobium]

Or try.

((test1<> 0x55) ) | ((test2<> 0xAA) )

[Chuck Hellebuyck]

I would have thought you would get an error but I believe there should be parenthesis around the whole IF operand.
i.e.
if ((test1<> 0x55)|(test2<> 0xAA))

[Marcoos]

I think it's just a bug, I tested all three variants without success, I'm testing it with Proteus and the generated code is obviously wrong.
What I seem a strange and that gives me doubts about my work is that writing a program, it is trivial to have such a condition, and it seems strange that it has not happened before.
For now I have taken the expression that I showed you working, but I think that must be corrected.
For now, thanks, Marco

[Anobium]

I have had the same experience last year.

Please post your full code (the code that does not work), what is the date of your GCB compiler?

[Anobium]

I an not sure that I understand your code. I have just run and get the results I expected, then, because I wanted to validate the results I ran the same tests in excel.

Four tests using the values in the first two columns.
With a calc of =IF(OR(A1<>85, B1<>170),"true","false"), where the data is in column A and B, the results are in column C
A B C
85 170 false
85 171 true
86 170 true
86 171 true

And, the results in GCB. The same.
false
true
true
true

With code as shown below changing the values for each loop.

test1 = 0x55
test2 = 0xAa

if ((test1<> 0x55) ) | ((test2<> 0xAA) ) THEN
HSerPrint "true"
else
HSerPrint "false"

END IF

Please post your full code (the code that does not work), what is the date of your GCB compiler?

[Marcoos]

in attached the two different test, with the generated assembler. You can see then whe I use the faulty mode the assembler execute a compmplement instruction, that is the error.
The compiler is the V9.30 22/09/2013
Marco

[Marcoos]

Hi
I have been worked to debug the problem, and the problem whith the pic compiler is not present, the assembler is quitly different, and work fine.
Meanwhile I have found another bug, due to an alignment error of my istruction, which exits the program without performing The Else.

;This not work but no error in compiling
if AppCountDown < MinCountDown Then
    Triac = 0 Else
    Triac = 1
end if

;This work fine
if AppCountDown < MinCountDown Then
    Triac = 0
Else
    Triac = 1
end if

I think that even in the first bug is an error in the management of complexity or managing the complexity of the line
Marco

[Anobium]

Please use your method that works.

To the best of my knowledge this is not supported. The latest help file has this information.

ELSE must be on a separate line in the source code.

Supported

....
<instruction> ' is supported

ELSE
<instruction>
.....

Not Supported

......
<instruction> ELSE ' is not supported but will compile 
<instruction> 
.....

[Anobium]

Regarding the compiler output from the different chips. Already spotted the difference when my code work (PIC) and your did not. :-)

Will respond when we have some news. I have added this the known issues list.

[Marcoos]

Anobium thanks,
I pointed out that I was writing my mistake, the problem is the failure to report the compiler! ;-)
I really like working with the GCB and I'm building a code generator for GCB state machine from one program to always open source
I will alert you as soon as usable
Marco