Date: Fri, 13 Mar 1998 17:49:42 +0100 From: lennartz.electronic@t-online.de (Thomas Zenker) To: freebsd-current@FreeBSD.ORG Subject: Re: A question about sys/sys/queue.h Message-ID: <19980313174942.53822@tue.le> In-Reply-To: <19980313121330.54903@follo.net>; from Eivind Eklund on Fri, Mar 13, 1998 at 12:13:30PM %2B0100 References: <XFMail.980312191745.shimon@simon-shapiro.org> <19980313121330.54903@follo.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, Mar 13, 1998 at 12:13:30PM +0100, Eivind Eklund wrote:
> On Thu, Mar 12, 1998 at 07:17:45PM -0800, Simon Shapiro wrote:
> > Why was the definition of some macros changed
> > from:
> >
> > #define FOO(a) { ... }
> >
> > to:
> >
> > #define FOO(a) do { ... } while(0)
> >
> > I thought these are the same...
>
> Imagine these used in a dual if () statement:
>
> if (bar)
> if (baz)
> FOO(1);
> else
> printf ("You loose!\n");
>
> With the former, you get something that (with proper indentation) map
> as
>
> if (bar)
> if (baz)
> { ... };
> else
> printf ("You loose!\n");
>
> while with the do {...} while (0) trick, you get
>
> if (bar)
> if (baz)
> do { ... } while(0);
> else
> printf ("You loose!\n");
>
It will not compile anyway.
The version w/ "{ ... };" is syntactically incorrect, because
there are two statements after the if, leaving the else w/o partner.
if (bar)
if (baz)
{ ... } /* compound_statement */
; /* empty but existing statment! */
else /* else without matching if! */
...
while the second version "do { ... } while (0);" still is a simple statement.
--
Thomas Zenker
at work thz@lennartz-electronic.de
private thz@tuebingen.netsurf.de
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19980313174942.53822>