From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1396 invoked from network); 7 Oct 1999 21:25:21 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 7 Oct 1999 21:25:21 -0000 Received: (qmail 21656 invoked by alias); 7 Oct 1999 21:25:15 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 8175 Received: (qmail 21649 invoked from network); 7 Oct 1999 21:25:14 -0000 To: zsh-workers@sunsite.auc.dk Subject: Reasons for not wanting EXTENDED_GLOB interactively (was Re: PATCH:...) References: From: Bruce Stephens Date: 07 Oct 1999 21:32:19 +0100 In-Reply-To: Zefram's message of "Thu, 7 Oct 1999 18:29:43 +0100 (BST)" Message-ID: <871zb6zzcc.fsf@cenderis.demon.co.uk> User-Agent: Gnus/5.070096 (Pterodactyl Gnus v0.96) XEmacs/21.1 (Biscayne) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Zefram writes: > I find it difficult to believe that anything actually relies on > "emulate zsh" *not* resetting EXTENDED_GLOB. This isn't really on the same subject, but a colleague wanted to delete files beginning .# in a directory a few days ago (they're created by Emacs ediffing files with versions or something), and did: rm .#* (Don't try this at home.) I'm not sure whether this is a bug in zsh. After all, the "rm *" warning is a special case (and useful, too, IMHO). Hmm, how about changing the implementation of the check? Presumably the current test looks for * explicitly; an alternative (slightly more expensive) would be to do the expansion and see if it includes all of the files or not, and if it does, issue the warning. In fact, one could probably do this as a shell function, instead. That probably makes more sense.