From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <zsh-workers-return-10477-mason-zsh=primenet.com.au@sunsite.auc.dk>
Received: (qmail 9671 invoked from network); 4 Apr 2000 17:17:54 -0000
Received: from sunsite.auc.dk (130.225.51.30)
  by ns1.primenet.com.au with SMTP; 4 Apr 2000 17:17:54 -0000
Received: (qmail 13243 invoked by alias); 4 Apr 2000 17:17:23 -0000
Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm
Precedence: bulk
X-No-Archive: yes
X-Seq: 10477
Received: (qmail 13207 invoked from network); 4 Apr 2000 17:17:21 -0000
Date: Tue, 4 Apr 2000 13:17:13 -0400
From: Clint Adams <schizo@debian.org>
To: zsh-workers@sunsite.auc.dk
Subject: PATCH: zfcd_match erratic on some awks
Message-ID: <20000404131713.A25031@dman.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
User-Agent: Mutt/1.1.2i

Seems that behavior for an index of 0 is undefined.  While this
works out fine for gawk and Solaris awk, mawk has different ideas
of what to do.  This should make everybody happy.

Index: Functions/Zftp/zfcd_match
===================================================================
RCS file: /cvsroot/zsh/zsh/Functions/Zftp/zfcd_match,v
retrieving revision 1.2
diff -u -r1.2 zfcd_match
--- Functions/Zftp/zfcd_match	2000/04/01 20:49:47	1.2
+++ Functions/Zftp/zfcd_match	2000/04/04 17:15:17
@@ -26,7 +26,7 @@
   # to list, but not a glob.  Don't ask me why.
   # I hate having to rely on awk here.
   zftp ls -LF $dir >$tmpf
-  reply=($(awk '/\/$/ { print substr($1, 0, length($1)-1) }' $tmpf))
+  reply=($(awk '/\/$/ { print substr($1, 1, length($1)-1) }' $tmpf))
   rm -f $tmpf
   [[ -n $dir && $dir != */ ]] && dir="$dir/"
   if [[ -n $WIDGET ]]; then