From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <zsh-workers-return-9524-mason-zsh=primenet.com.au@sunsite.auc.dk>
Received: (qmail 11480 invoked from network); 2 Feb 2000 08:45:59 -0000
Received: from sunsite.auc.dk (130.225.51.30)
  by ns1.primenet.com.au with SMTP; 2 Feb 2000 08:45:59 -0000
Received: (qmail 13201 invoked by alias); 2 Feb 2000 08:45:55 -0000
Mailing-List: contact zsh-workers-help@sunsite.auc.dk; run by ezmlm
Precedence: bulk
X-No-Archive: yes
X-Seq: 9524
Received: (qmail 13193 invoked from network); 2 Feb 2000 08:45:54 -0000
Date: Wed, 2 Feb 2000 09:45:53 +0100 (MET)
Message-Id: <200002020845.JAA09989@beta.informatik.hu-berlin.de>
From: Sven Wischnowsky <wischnow@informatik.hu-berlin.de>
To: zsh-workers@sunsite.auc.dk
In-reply-to: Peter Stephenson's message of Tue, 01 Feb 2000 20:53:26 +0000
Subject: Re: Wordcode bug with $functions


Peter Stephenson wrote:

> I'm getting a segmentation fault with `print $functions', even patched up
> to date (supposedly). I presume this is a new wordcode feature.  This is
> reproducible with my addpath function by doing `which addpath'.  I narrowed
> it down:  actually, this form doesn't give a crash, but it does show more
> clearly what's going wrong.

When dumping function definitions it didn't set state->pc to the
calculated end.

> By the way, I found this by doing `set' on its own.  This dumps out the
> whole of every function when zsh/parameter is loaded.  Yuck.

And usernames and commands and ... But that's how it's supposed to
work, isn't it? And it was rather ugly before already (e.g. $_comps
and probably other parameters like the caches).

Bye
 Sven

diff -ru ../z.old/Src/text.c Src/text.c
--- ../z.old/Src/text.c	Wed Feb  2 09:34:48 2000
+++ Src/text.c	Wed Feb  2 09:40:40 2000
@@ -182,6 +182,7 @@
 	} _redir;
 	struct {
 	    char *strs;
+	    Wordcode end;
 	} _funcdef;
 	struct {
 	    Wordcode end;
@@ -372,11 +373,13 @@
 		    taddnl();
 		    n = tpush(code, 1);
 		    n->u._funcdef.strs = state->strs;
+		    n->u._funcdef.end = end;
 		    state->strs = (char *) (p + (*state->pc));
 		    state->pc += 2;
 		}
 	    } else {
 		state->strs = s->u._funcdef.strs;
+		state->pc = s->u._funcdef.end;
 		tindent--;
 		taddnl();
 		taddstr("}");

--
Sven Wischnowsky                         wischnow@informatik.hu-berlin.de