From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 10801 invoked from network); 15 Sep 1999 14:16:33 -0000 Received: from sunsite.auc.dk (130.225.51.30) by ns1.primenet.com.au with SMTP; 15 Sep 1999 14:16:33 -0000 Received: (qmail 3798 invoked by alias); 15 Sep 1999 14:16:23 -0000 Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm Precedence: bulk X-No-Archive: yes X-Seq: 7847 Received: (qmail 3790 invoked from network); 15 Sep 1999 14:16:20 -0000 Date: Wed, 15 Sep 1999 16:16:18 +0200 (MET DST) Message-Id: <199909151416.QAA32682@beta.informatik.hu-berlin.de> From: Sven Wischnowsky To: zsh-workers@sunsite.auc.dk In-reply-to: Peter Stephenson's message of Wed, 15 Sep 1999 15:05:32 +0200 Subject: Re: PATCH: 3.1.6-pws-4: backreferences Peter Stephenson wrote: > Tanaka Akira wrote: > > Z(2):akr@is27e1u11% Src/zsh -f > > is27e1u11% setopt extendedglob > > is27e1u11% a=abcdef > > is27e1u11% echo ${a/(#b)a(?)c/$match[1]} > > def > > > > It should be `bdef'. Since second time, it works well. > > I've tried this on three different systems, with all the latest patches, > and I don't get this, so I'm stuck at the moment. I've run out of sync, then, because I get it too. And no wonder, in subst.c:1291 replstr is still unconditionally singsub()ed (and indeed if I execute different expansions, I always get the previous $match[1] inserted). And even if I quote the $match[1], it will not be singsub()ed again (or not correctly? there is no tokenize() before the singusb() at glob.c:1850). Hm, so how is this supposed to work: with or without quoting the beast? Bye Sven -- Sven Wischnowsky wischnow@informatik.hu-berlin.de