PureBasic Forums - English

Lewis wrote:Sure, although this doesn't contribute much to the TRUE/FALSE thread (for what that academic conversation is worth).

@Lewis: The thread topic actually relates to what PB variable type represents a boolean. The answers which were given were that byte, word, long, quad, and integer could be used.

The academic part, which I believe was started by you, doesn't really relate to the thread. I admit I departed from the thread topic a little when I was responding to the ideas you stated as being important.

:roll: Very observant of you, blueznl! However, the value not-zero is impossible to use directly in calculations, unlike a TRUE value of -1, which in certain circumstances can also improve code readability. Tips: Code and run X = NOT 0, and then code and run Y = NOT -1.

I found several incompatibilities in your assumptions. The first is that IMHO Boolean types are primarily used for conditionals and not calculations. Using a Boolean type in a calculation is more a characteristic of slow computers or limited memory than sound programming strategy (though it does have its uses ). Second, if you try to use a Boolean type as part of a calculation it is possible to do it with TRUE being defined as either -1 or as 1 (see my previous code sample). Third, if you are coding something that uses Boolean types in calculations it is your responsibility for making sure that the dual-use of the Boolean type (as both a logic result and as a numeric type) is aligned with the way you are using it.


Fred said that he planned on making a Boolean() function that would evaluate a conditional expression and would be usable in calculations. I'm not sure what it would return for the TRUE value. As you stated, it would be important that it returns a consistent value, i.e. -1 or 1 but not both, whenever it evaluates to TRUE.

@Demivec: You seem to be missing the point! Choosing the TRUE value of 1 (or any value other than 0) and using it in calculations requires that you make this known for later reading of the code (by yourself or others). Using the true TRUE value of -1 would not require further documentation -- provided it was implemented by the programming language. That said, I've stated that my views are to be treated purely academically since PB is highly unlikely to implement them so far into its development (too many programs would need to be checked and possibly changed).

Cheers,
Lewis

PS. If you insist on getting in the last word, go ahead...

Lewis wrote:@Demivec: You seem to be missing the point! Choosing the TRUE value of 1 (or any value other than 0) and using it in calculations requires that you make this known for later reading of the code (by yourself or others). Using the true TRUE value of -1 would not require further documentation -- provided it was implemented by the programming language. That said, I've stated that my views are to be treated purely academically since PB is highly unlikely to implement them so far into its development (too many programs would need to be checked and possibly changed).

Cheers,
Lewis

PS. If you insist on getting in the lasting word, go ahead...

Hardly takes rocket science does it?

@LuCiFeR[SD]: Nice contribution, however, work-arounds are just that -- work-arounds.

Cheers,
Lewis

Lewis wrote:PS. If you insist on getting in the last word, go ahead...

It's not a workaround though. It does exactly what is required. You could easily pull the line out of the macro and use it directly in the code.

all you need is the
The only difference is the macro, which could just as easily be a procedure, improves readability. as the name just tells you what it does. use a few comments here and there to help you out and all is good with the world, no hacks, no bodges, just simple code.