Does PB short-circuit expressions?

matthew180 · Post by **matthew180** » Wed Jun 30, 2004 7:52 pm

Greetings,

The subject pretty much sums it up... Does PB perform short-circuiting or expressions?

Thanks,
Matthew

Karbon · Post by **Karbon** » Wed Jun 30, 2004 8:48 pm

Matthew. I already told you.

YES.

Don't make me come up there.

matthew180 · Post by **matthew180** » Wed Jun 30, 2004 9:00 pm

SILENCE!

It's not specifically stated in the manual and I didn't feel like writing a test. However, I did write a test program anyway and you are correct, it does do short circuit. But I also wanted the question here in the forum for other people to find when they search (I'm sure the question will come up again.)

M@

Kale · Post by **Kale** » Wed Jun 30, 2004 9:04 pm

Does PB perform short-circuiting or expressions?

Whats the difference? *starts digging out computer science books*

matthew180 · Post by **matthew180** » Wed Jun 30, 2004 9:44 pm

Oops, that's supposed to be on not or...

I probably should have asked something more along the lines of:

Does PB do short-circuit during the evaluation of expressions?

Sorry for the confusion!

M@

Kale · Post by **Kale** » Thu Jul 01, 2004 12:10 am

ah, i see, but still explain short circuiting please.

Post by **Dare2** » Thu Jul 01, 2004 1:29 am

[EDIT]

This post deleted because the information herein was erroneous.

Grrr. I felt like saying: This post intentionally left blank. This is a true reflection of the IQ level of the poster.

tinman · Post by **tinman** » Thu Jul 01, 2004 8:42 am

[quote="Dare2]A quick example cure:[/quote]

Short circuit evaluation is a good thing. It means you don't need to evaluate both sides of a logical expression to get the result. Therefore if you have a simple comparison logically combined with an evil complex one, you're best putting the simple one on the left because it may mean you can avoid carrying out the right hand side thus saving you time.

Conversely, if you have e.g. a procedure which must always be evaluated, you should put it on the left.

The topic you (IIRC) started seems to indicate a bug.

GedB · Post by **GedB** » Thu Jul 01, 2004 9:40 am

Short circuiting allows for some neat tricks:

viewtopic.php?t=9926&highlight=die

The thing to remember is that when you have exp1 OR exp2, exp2 will only get run if exp1 returns fales. This can be useful if you want to handle the expception.

Fred advices against the use as I describe it in the above post. Using expressions as commands can have unexpected results.

The way I do it now is:

Code: Select all

If OpenWindow(...) or Die("Failed to Open Window)
    If If CreateGadgetList(WindowID()) or Die("Failed to Create Gadget List")
       [...]
    EndIf
EndIf

The calls to Die only get made if the first function returns 0.

With And the opposite is true. The second expression only gets executed if the first is successful. This is useful for dependancies. For example:

Code: Select all

If InitSprite() And OpenScreen(x1, y1, 16, "Title")
   [...]
EndIf

OpenScreen is only called if InitSprite is successful. I think this is tidier than nesting Ifs. More complex expressions are possible, such as:

Code: Select all

If ( OpenWindow(...) and CreateGadgetList(WindowID() ) or Die("Window Error")
       [...]
EndIf

You just have to avoid making it incomprehensible

Here's a working sample to cut and paste:

Code: Select all

Procedure Die(Message.s)
  MessageRequester("Fatal Error", Message)
  End
EndProcedure

#Flags = #PB_Window_WindowCentered | #PB_Window_SystemMenu
If (OpenWindow(0, 0, 0, 100, 100,#Flags , "Test") And CreateGadgetList(WindowID())) Or Die("Window Error")
  EditorGadget(0, 0, 0, 100, 100)
EndIf

Repeat
  evt = WaitWindowEvent()
Until evt = #PB_Event_CloseWindow

PB · Post by PB » Thu Jul 01, 2004 10:39 am

> Short circuit evaluation is a good thing. It means you don't need to
> evaluate both sides of a logical expression to get the result.

True, and I like that aspect of PureBasic. I vote to keep it that way, because
I can always just use brackets if I ever need to force an equation.

Kale · Post by **Kale** » Thu Jul 01, 2004 11:22 am

I've just realised i do understand 'short circuiting' and why its a good thing, i've just never heard it called that before.

Post by **Dare2** » Thu Jul 01, 2004 11:30 am

@tinman

Did I get the whole thing wrong? (again)

tinman · Post by **tinman** » Thu Jul 01, 2004 12:13 pm

Dare2 wrote:@tinman

Did I get the whole thing wrong? (again) :?

Just that short circuit evaluation in itself isn't a bug or problem, but it does seem to be broken at the moment.

Post by **Dare2** » Thu Jul 01, 2004 12:31 pm

Okay.

Karbon · Post by **Karbon** » Thu Jul 01, 2004 12:51 pm

MAtthew is great at breaking code.. He's been breaking mine for years