Announcement

**Patrice Terrier** · 25 Sep 2003, 10:41 PM

Paul,

Try this:

True& = NOT True&
PRINT True& ' -1
True& = NOT True&
PRINT True& ' 0

this is the reason why I like using True& = -1 myself
because it works well with the NOT operator
then I don't need the overhead of functions like ISTRUE/ISFALSE.

%TRUE = 1 is the C definition,
%TRUE = -1 has always been the "TRUE" BASIC definition.

That's my point

------------------
Patrice Terrier
mailto

[email protected][email protected]</A>
http://www.zapsolution.com
Toolkit: WinLIFT (Skin Engine), GDI+ helper (Graphic package), dvBTree (Index manager)
Freeware: ZAP Audio Player, ZAP Picture Browser.
Artwork on demand.

**Paul Dwyer** · 25 Sep 2003, 11:13 PM

hmmmmm.

Any annoying situation though having two types. Generally I don't like to think about the actual values of true and false but just use them. Problem is that if I use:

If MyFunc() = %true then

I may never execute the code if it's someone elses as they are passing me a -1.

In the case of Clay's code I'm glad he qualified what he meant by True as often people don't.

I'm a big fan of standards thats all... don't get me started about DVD+R and DVD-R

------------------
Paul Dwyer
Network Engineer
Aussie in Tokyo

[This message has been edited by Paul Dwyer (edited September 26, 2003).]

**Elias Montoya** · 26 Sep 2003, 12:01 AM

hehehe

I remember in the days of Qbasic...

CONST TRUE = -1
CONST FALSE = NOT TRUE

------------------
Learning is not the capacity to retain information, It is the capacity to compare old information with new information... And make new Knowledge.

**Paul Dwyer** · 26 Sep 2003, 12:17 AM

Well,

This works with the %true/%false I use, so I suppose I'm happy....

Code:

#Compile Exe
 
#Include "win32api.inc" 

Function PbMain() As Long
    
    Dim T As Long
    Dim F As Long
    
    T = %true
    F = %false
    
    If IsTrue(T) Then
        MsgBox "T is True"
    Else
        MsgBox "T is Not True"
    End If
    
    If IsFalse(F) Then
        MsgBox "F is False"
    Else
        MsgBox "F is Not False"
    End If
    
End Function

------------------
Paul Dwyer
Network Engineer
Aussie in Tokyo

**Dale Yarker** · 26 Sep 2003, 01:25 AM

I got used to doing "IF ISTRUE ...", and "IF ISFALSE ...", instead of
"IF x = %true" etc. ISTRUE and ISFALSE may have been invented to avoid
confusion with NOT in IF statements, they also make this discussion
redundant.

FALSE = zero; TRUE = anything else, set it to what you like, doesn't
matter. And of course, NOT FALSE = -1& and NOT FALSE = 4,294,967,295???

Cheers guys,

------------------
Dale
mailto:[email protected][email protected]</A>

**Knuth Konrad** · 26 Sep 2003, 01:28 AM

I remember reading that FALSE is defined as 0 and TRUE as NOT FALSE. So it really doesn't matter if TRUE = 1 or TRUE = -1. Both is NOT FALSE.

At least PB's IsTrue() and IsFalse() work that way.

Knuth

------------------
http://www.softAware.de

**Mike Trader** · 26 Sep 2003, 02:13 AM

Exactly,

FALSE = 0
TRUE = everything else = NOT FALSE

I actually like this cos I use

IF SomeFlag then

I can then use 1 for sub condition 1, 2 for sub condition 2 etc
or 1 and -1 for long and short ie

If TradeFlag Then ' in a trade
If TradeFlag < 0 Then ' Short Processing

If TradeFlag > 0 Then ' Long Processing

------------------
Kind Regards
Mike

**Edwin Knoppert** · 26 Sep 2003, 02:22 AM

I never make use of these constants.
I just use 0 and 1 since a 'boolean' this way can be upgraded to a number later like:
0,1,2 etc..

------------------
http://www.hellobasic.com

**Eric Pearson** · 26 Sep 2003, 02:43 AM

I too have a strong preference for %FALSE = 0 and %TRUE = -1 because those values work so well with the logical operators. For example using -1 allows me to use NOT in a logical way, as in...

Code:

IF NOT lUserAbort THEN...

But complete standardization is impossible because not all data types can hold negative numbers. WORDs, DWORDs, BYTEs, etc. can't store negative one.

The problem is further complicated by the fact that if you set all of the bits of a BYTE you get 255, if you set all of the bits of a WORD you get 64k, and so on. So "all bits on" doesn't work as a standard either. Every data type would have a different value for True. Not good.

So Microsoft settled on positive one, which all data types can store. It's really the only value that makes sense as a "universal true". The %TRUE equate in W=in32API.INC is a Microsoft value, not a PowerBASIC-defined value. When Win32.HLP and MSDN say "True" they almost always mean positive one.

If you want to use signed integers such as LONGs for your logical values, then you are free to re-define %TRUE as negative one, as I do. I change Win32API.INC to...

Code:

%FALSE    = 0
%TRUE     = -1
%API_TRUE = +1

In the end, I hardly ever use %API_TRUE for anything.

-- Eric

------------------
Perfect Sync Development Tools
Perfect Sync Web Site
Contact Us: mailto:[email protected][email protected]</A>

[This message has been edited by Eric Pearson (edited September 26, 2003).]

**Nathan Evans** · 26 Sep 2003, 03:55 AM

False = Zero
True = non-Zero

------------------

**Scott Slater** · 26 Sep 2003, 04:45 AM

It actually has to do with the BINARY values given. A logical NOT looks for
the OPOSITE value. 1 is stored as "0000000000000001" whereas -1 is stored as
"1111111111111111" So NOT -1 would translate to "0000000000000000", and NOT 1
would translate to "1111111111111110" which is still a "non-zero" value (-2).

------------------
Scott Slater
Summit Computer Networks, Inc.
www.summitcn.com

**Don Schullian** · 26 Sep 2003, 06:13 AM

Ever since Commodor days I've NEVER trusted Ture!
False is ZERO, always, no exceptions!

True, therefore is NOT ZERO. eg:

IF Whatever <> 0 THEN
PRINT "This is ALWAYS a good test for NOT ZERO!"
END IF

------------------
C'ya
Don
[email protected]

**Paul Dwyer** · 26 Sep 2003, 08:08 AM

For those of you who are from a VB background you will recall one more little annoyance though,

Many functions in the language returned -1 as an error. So if you had this as TRUE you would run into problems

This was sometimes sensible too as you could do what Mike was mentioning

If MyFunt() then

but if -1 is true .....

As the topic states, this bothers me somewhat, especially when having to support other people's code

------------------
Paul Dwyer
Network Engineer
Aussie in Tokyo

**Keith Waters** · 26 Sep 2003, 09:30 AM

PB should have a proper intrinsic BOOLEAN type (is it already on the wish list?). This would avoid having to use an integer type, with all the consequent confusion and scope for error. Mind you, I can see that this would be a bit difficult to add now without introducing backward compatibility problems.

Keith

------------------

**Mike Trader** · 26 Sep 2003, 11:58 AM

Yeah, what he said ...

------------------
Kind Regards
Mike

**Gary Peek** · 26 Sep 2003, 01:47 PM

Paul Dwyer wrote: Why can't we agree on TRUE and FALSE?

I just say, "@#$% 'em both!"

I use 0 and 1. That way I know what the @#$% I have.
Who wants to type that %TRUE and %FALSE junk anyway.

(I admit there are uses for them, but most of the time
I don't need it.)

------------------

**Michael Mattias** · 26 Sep 2003, 03:52 PM

Why can't we agree on TRUE and FALSE?

Better idea: Why can't we just faggedaboudit?

When in Rome....use ISTRUE and ISFALSE.

MCM

Announcement

Grrrrrr :-( Does this bother anyone else?

Grrrrrr :-( Does this bother anyone else?

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment