From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 12832 invoked from network); 8 May 2000 10:41:25 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 8 May 2000 10:41:25 -0000 Received: (qmail 666 invoked by alias); 8 May 2000 10:41:04 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 11254 Received: (qmail 556 invoked from network); 8 May 2000 10:40:56 -0000 Date: Mon, 8 May 2000 12:40:49 +0200 (MET DST) Message-Id: <200005081040.MAA20868@beta.informatik.hu-berlin.de> From: Sven Wischnowsky To: zsh-workers@sunsite.auc.dk In-reply-to: "Andrej Borsenkow"'s message of Fri, 5 May 2000 18:53:49 +0400 Subject: RE: Command completion in nslookup Andrej Borsenkow wrote: > [ Bart wrote: ] > > really-do-complete-word-even-in-column-one() { zle > > complete-word "$@" } > > bindkey '\t' really-do-complete-word-even-in-column-one > > But how do I get it for nslookup only (or for any nslookup-like case)? > I'd prefer some context-based solution ... Sven's suggestion looks good > with some shell-level support. Hm, what I had in mind wouldn't have allowed us to configure it. Another suggestion: allow compstate[insert] to take the value `tab' (or something like this). On entry to the widget it has this value when in the first column. On exit, if it has this value, only a T=tab will be inserted. But then the widget can decide what it wants to do and we could easily add a style for this. Good? Or is this too weird? (Although I find that quite consistent.) Bye Sven -- Sven Wischnowsky wischnow@informatik.hu-berlin.de