Date: Wed, 2 Jan 2002 09:34:05 -0800 (PST) From: Julian Elischer <julian@elischer.org> To: Stephen McKay <mckay@thehub.com.au> Cc: John Baldwin <jhb@FreeBSD.org>, Poul-Henning Kamp <phk@critter.freebsd.dk>, cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org, Julian Elischer <julian@FreeBSD.org>, Greg Lehey <grog@FreeBSD.org> Subject: Re: cvs commit: src/share/man/man3 queue.3 Message-ID: <Pine.BSF.4.21.0201020933320.19204-100000@InterJet.elischer.org> In-Reply-To: <200201021440.g02Eess11512@dungeon.home>
next in thread | previous in thread | raw e-mail | index | archive | help
how about explicitly detailing that it's NULL at the end?
Doesn't that also work?
On Thu, 3 Jan 2002, Stephen McKay wrote:
> On Tuesday, 1st January 2002, Julian Elischer wrote:
>
> >> > TAILQ_FOREACH(p, &list, p_next)
> >
> >So you think that the final value of p should be undefined,
> >and there is no built in way to see if there were any elements in the
> >list at all or whether we completed the list?
>
> I'd have to say, "Yes". I expect p to be undefined when the TAILQ_FOREACH
> completes. It's an internal implementation detail. Similarly, I wouldn't
> depend on p == head for CIRCLEQ_FOREACH. You can roll your own loop if
> you want special behaviour.
>
> >I'm just documenting what the code does, and how people have used it.
> >if you think that phk is correct we should add an example to the man page
> >showing:
>
> Perhaps Marcel's good example instead:
>
> > TAILQ_FOREACH(p, &list, p_next) {
> > if (some_test_on(p)) {
> > /* Do something */
> > break; /* optional */
> > }
> > }
>
> The man page should explicitly state that p is undefined after the loop
> completes unless the user breaks out. Oh, and explicitly stating that it
> is OK to break out at all might be nice too. :-)
>
> Stephen.
>
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0201020933320.19204-100000>