From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23347 invoked from network); 12 Nov 1998 10:07:18 -0000 Received: from math.gatech.edu (list@130.207.146.50) by ns1.primenet.com.au with SMTP; 12 Nov 1998 10:07:18 -0000 Received: (from list@localhost) by math.gatech.edu (8.9.1/8.9.1) id FAA23658; Thu, 12 Nov 1998 05:06:32 -0500 (EST) Resent-Date: Thu, 12 Nov 1998 05:06:32 -0500 (EST) From: "Bart Schaefer" Message-Id: <981112020422.ZM10056@candle.brasslantern.com> Date: Thu, 12 Nov 1998 02:04:22 -0800 In-Reply-To: <19981112172055.A21356@primenet.com.au> Comments: In reply to Geoff Wing "PATCH: 3.1.5: ``***'' symlink follow broken" (Nov 12, 5:20pm) References: <19981112172055.A21356@primenet.com.au> X-Mailer: Z-Mail (4.0b.820 20aug96) To: Geoff Wing , Zsh Hackers Subject: Re: PATCH: 3.1.5: ``***'' symlink follow broken MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Resent-Message-ID: <"0_MTA3.0.bn5.eChIs"@math> Resent-From: zsh-workers@math.gatech.edu X-Mailing-List: archive/latest/4611 X-Loop: zsh-workers@math.gatech.edu Precedence: list Resent-Sender: zsh-workers-request@math.gatech.edu On Nov 12, 5:20pm, Geoff Wing wrote: } Subject: PATCH: 3.1.5: ``***'' symlink follow broken } } ``***'' recursive globbing with symlink follow has been broken since at } least 3.1.2 (maybe before): [...] } This patch is probably suboptimal (possibly wrong) and may encourage someone } to make a real patch. [...] } ! if (dirs && !q->follow) { I think the semantically correct test is if (dirs && !closure) { but it may be that the only time it makes a difference is when all of the subdirectories are symlinks. I keep thinking there must be some other case that will fail if !q->follow is tested instead, but I can't quite figure out what. -- Bart Schaefer Brass Lantern Enterprises http://www.well.com/user/barts http://www.brasslantern.com