From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1717 invoked from network); 20 Oct 1998 17:45:12 -0000 Received: from math.gatech.edu (list@130.207.146.50) by ns1.primenet.com.au with SMTP; 20 Oct 1998 17:45:12 -0000 Received: (from list@localhost) by math.gatech.edu (8.9.1/8.9.1) id NAA05033; Tue, 20 Oct 1998 13:36:48 -0400 (EDT) Resent-Date: Tue, 20 Oct 1998 13:36:28 -0400 (EDT) From: "Bart Schaefer" Message-Id: <981020095542.ZM5494@candle.brasslantern.com> Date: Tue, 20 Oct 1998 09:55:42 -0700 In-Reply-To: <362C9FB4.B4C08045@guildford.ericsson.se> Comments: In reply to Daniel Rigal "Backticks In zshenv" (Oct 20, 3:35pm) References: <362C9FB4.B4C08045@guildford.ericsson.se> X-Mailer: Z-Mail (4.0b.820 20aug96) To: Daniel Rigal , zsh-users@math.gatech.edu Subject: Re: Backticks In zshenv MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Resent-Message-ID: <"Ryqvs2.0.0E1.SeCBs"@math> Resent-From: zsh-users@math.gatech.edu X-Mailing-List: archive/latest/1877 X-Loop: zsh-users@math.gatech.edu X-Loop: zsh-workers@math.gatech.edu Precedence: list Resent-Sender: zsh-workers-request@math.gatech.edu On Oct 20, 3:35pm, Daniel Rigal wrote: } Subject: Backticks In zshenv } } I am having a problem which is driving me up the wall! } Basically, Z-Shell seems to randomly decline to run commands } in backticks in the zshenv (both the central one or the } user's own zshenv show this effect). I've just seen the same problem, on a newly-installed RedHat 5.1 Linux system with the zsh-3.0.5-2 RPM. The /etc/zshenv file contains if [[ `id -u` -eq 0 ]] then ... fi and zsh randomly fails to capture the output of `id -u`. It seems to occur only when I rapidly exit and immediately log back in; if I wait any longer than it takes to retype my login and password as fast as I can, the problem disappears or becomes much more sporadic. I ended up replacing `id -u` with $EUID and haven't seen any further problems. I'd assumed it was something about the "id" program. I suspect zsh is getting some kind of a signal -- perhaps SIGWINCH? -- which is interrupting the read on the command substitution. -- Bart Schaefer Brass Lantern Enterprises http://www.well.com/user/barts http://www.brasslantern.com