$IF NOT is not?

[Daniel Raymer]

$IF NOT....$ENDIF is given in the PB help listing but not in my old manual. It does not seem to work, and just compiles it regardless of the parameter value. But, it doesn't give an error when you compile.

Try this, and change %RDSpro from 0 to 1:
------------------------------------------
cls
%RDSpro=0 'both if 1 only ifnot if 0
$IF %RDSpro
PRINT " if "
$ENDIF
$IF NOT %RDSpro
PRINT " if not"
$ENDIF
end
-----------------------------------------

It should toggle, showing one or the other. Instead you get both if =1, and only "if not" if =0

[Daniel Raymer]

obvious workaround

BTW, the obvious workaround is to use:

$IF %RDSpro
$ELSE
PRINT " if not"
$ENDIF

...but that is not very elegant now, is it?

So, is $IF NOT supposed to work in PB-DOS?

[Mel Bishop]

It's been a while since I played with PB/DOS but you can try....

1. Dump the '$' in front of all IF statements.
2. Change ENDIF to END IF (with a space).

I'm probably wrong about this but I think '$' denotes a compiler metastatement. The code will be dealt with at compile time, not at run time.

[Daniel Raymer]

thanks but...

It is a compiler metastatement. It tells the compiler to skip lines of code if the constant is true (>0).

$IF...$ENDIF works just fine.

[Michael Mattias]

'NOT 1' does not equal zero.

[Daniel Raymer]

not

But I thought that not_true = false.

Philosophy wasn't my best subject.

So, my dumb workaround is probably the smart way to go.

[Eric Pearson]

Dan --

Michael is correct, NOT 1 is not FALSE.

If you use 0 and -1 (instead of 1) then $IF NOT will work the way you expect.

-- Eric

[Mark Hunter]

Basic, Power or otherwise, uses the constants 0 for False and –1 for True.

Not n takes n in binary and reverses all the bits. Since –1 is represented as the 16 bit word 1111111111111111 this means Not n = –1 – n. The following PBDOS program illustrates this (leading 0s aren’t shown):

 For n = -10 To 10
  Print Not n; -1 - n, Bin$(n), Bin$(-1 - n)
 Next

[Michael Mattias]

%FALSE = 0
%TRUE  = NOT %FALSE

%AnyEquate =  %TRUE | %FALSE
....

$IF [NOT] %Anyequate
...
$ELSE
...
$ENDIF

[Dave Biggs]

%TRUE = NOT %FALSE

That's incorrect -

Win32API.inc defines the equates
%TRUE = 1
%FALSE = 0

Therfore Not %FALSE = -1. Which is True but isn't %TRUE

[Michael Mattias]

Win32API.inc defines the equates
%TRUE = 1
%FALSE = 0

This is the PB/DOS forum where Win32API.INC holds no sway.

Regardless...my solution works, don't it?

[Michael Mattias]

Win32API.inc defines the equates
%TRUE = 1
%FALSE = 0

Just for the record, as long as you defined %TRUE as any non-zero integer, your Windows' programs will work correctly.... assuming you coded them correctly to begin with, using ISTRUE and ISFALSE instead of testing numeric values versus either %TRUE or %FALSE.

[Dave Biggs]

I should think that using ISTRUE and ISFALSE would be a better way to test the logical truth or falsity of an expression even in PBDos.

[Arthur Gomide]

Daniel,

If you use the regular Basic values of TRUE = -1 and FALSE = 0, your code works perfectly!