From owner-cvs-all  Wed Jan  2  0:41:16 2002
Delivered-To: cvs-all@freebsd.org
Received: from mailhub.fokus.gmd.de (mailhub.fokus.gmd.de [193.174.154.14])
	by hub.freebsd.org (Postfix) with ESMTP
	id F08C837B41B; Wed,  2 Jan 2002 00:41:07 -0800 (PST)
Received: from beagle (beagle [193.175.132.100])
	by mailhub.fokus.gmd.de (8.11.6/8.11.6) with ESMTP id g028f0h00429;
	Wed, 2 Jan 2002 09:41:03 +0100 (MET)
Date: Wed, 2 Jan 2002 09:41:00 +0100 (CET)
From: Harti Brandt <brandt@fokus.gmd.de>
To: Marcel Moolenaar <marcel@xcllnt.net>
Cc: cvs-all@FreeBSD.ORG, <cvs-committers@FreeBSD.ORG>
Subject: Re: cvs commit: src/share/man/man3 queue.3
In-Reply-To: <20011231190126.A755@dhcp01.pn.xcllnt.net>
Message-ID: <20020102093537.X1528-100000@beagle.fokus.gmd.de>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk
List-ID: <cvs-all.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20cvs-all>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20cvs-all>
X-Loop: FreeBSD.ORG

On Mon, 31 Dec 2001, Marcel Moolenaar wrote:

MM>On Mon, Dec 31, 2001 at 05:14:08PM -0800, Julian Elischer wrote:
MM>>
MM>> > I'd consider this a misuse of _FOREACH, because the something that's
MM>> > being done is definitely not done for each p. Even some_test_on() is
MM>> > not performed on each p. I'm with phk here.
MM>> >
MM>> > Note that if it's simply too easy to use _FOREACH in these cases, we
MM>> > may want to consider creating a more suitable abstraction, like:
MM>> >
MM>> > 	TAILQ_UNTIL(p, &list, p_next)
MM>> > 		some_test_on(p);
MM>> > 	if (p == 0)
MM>> > 		return ENOTFOUND;
MM>> > 	/* do something */
MM>> > 	return 0;
MM>> >
MM>> > Just a thought.
MM>>
MM>> So you think that the final value of p should be undefined,
MM>> and there is no  built in way to see if there were any elements in the
MM>> list at all or whether we completed the list?
MM>
MM>Theoretically speaking, yes.
MM>
MM>> I'm just documenting what the code does, and how people have used it.
MM>
MM>I understand that and didn't really object to that.
MM>
MM>> if you think that phk is correct we should add an example to the man page
MM>> showing:
MM>>
MM>>
MM>> TAILQ_FOREACH(p, &list, p_next)
MM>> 	if (some_test_on(p))
MM>> 		goto fred;
MM>> 	goto jim;
MM>> fred:	print ("we got one");
MM>> jim:
MM>
MM>What's wrong with the following if _FOREACH is preferred:
MM>
MM>	TAILQ_FOREACH(p, &list, p_next) {
MM>		if (some_test_on(p)) {
MM>			/* Do something */
MM>			break; /* optional */
MM>		}
MM>	}

This makes things harder to read if /* Do something */ is longer than a
screen. Well, I think, that relying on the pointer beeing NULL if running
to the end of the list just perfectly mirrors the usual C-idiom:

	for (i = 0; i < n; i++)
		if (test(i))
			break;

	if (i == n)
		...

The TAILQ_UNTIL(...) would even more obscure things...

harti
-- 
harti brandt, http://www.fokus.gmd.de/research/cc/cats/employees/hartmut.brandt/private
              brandt@fokus.gmd.de


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message