Furthermore:
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vclang/html
/_clang_precedence_and_order_of_evaluation.asp

yes, ms follows standards.


omega
Blackened Interactive - http://www.blackened-interactive.com
Wavelength - http://www.thewavelength.net

> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:hlcoders-
> [EMAIL PROTECTED] On Behalf Of Persuter
> Sent: June 8, 2003 11:52 PM
> To: [EMAIL PROTECTED]
> Subject: Re: [hlcoders] SDK line crashing linux servers
>
> At 03:32 PM 6/9/2003 +1200, you wrote:
>
>
> >>I'd say the most likely explanation is indeed shimms' and hofer's
> >>conjecture, i.e., that it has not been initialized. There is no serious
> >>compiler that will not follow the left-to-right rule, that's a fairly
> >>integral part of the C++ specification, which is used a lot (it's used
> >>throughout the HL SDK, for example), and certainly they all follow the
> >>correct precedence rules.
> >>
> >>Persuter
> >>--
> >
> >
> >I think it can depend on whether the compiler would use 'full' boolean
> >logic (where all of a statement is evaluated, even if it's something
> simple
> >like 1 && 0), or 'short circuit' (where the code will not evaluate
> further
> >if it's obvious what the result will be, like 0 || 0) logic... I know
> >Delphi (Object Pascal) can use either one, I'm sure C++ ones can too.
> >
> >As for sticking to a standard, since when has MS started doing that?
>
> Siiiiigh...
>
> C++ uses lazy ("short circuit"), left-to-right evaluation. All serious C++
> compilers follow that. Seriously, I can't believe none of you have noticed
> this, it is used quite a bit in the HL SDK.
>
> By the way, 0 || 0 is not amenable to short circuit evaluation, since the
> first one is not true. I think you meant 0 && 0. And obviously if you
> actually put 1 || 0 the compiler will go ahead and evaluate that at
> compile-time to true.
>
> Persuter
>
> _______________________________________________
> To unsubscribe, edit your list preferences, or view the list archives,
> please visit:
> http://list.valvesoftware.com/mailman/listinfo/hlcoders
>


_______________________________________________
To unsubscribe, edit your list preferences, or view the list archives, please visit:
http://list.valvesoftware.com/mailman/listinfo/hlcoders

Reply via email to