From owner-freebsd-ports  Mon Jul 17 01:30:27 1995
Return-Path: ports-owner
Received: (from majordom@localhost)
          by freefall.cdrom.com (8.6.10/8.6.6) id BAA13143
          for ports-outgoing; Mon, 17 Jul 1995 01:30:27 -0700
Received: from ghpc6.ihf.rwth-aachen.de (ghpc6.ihf.RWTH-Aachen.DE [134.130.90.6])
          by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id BAA13130
          for <ports@freebsd.org>; Mon, 17 Jul 1995 01:30:20 -0700
Received: (from thomas@localhost) by ghpc6.ihf.rwth-aachen.de (8.6.11/8.6.9) id KAA28843; Mon, 17 Jul 1995 10:30:02 +0200
From: Thomas Gellekum <thomas@ghpc8.ihf.rwth-aachen.de>
Message-Id: <199507170830.KAA28843@ghpc6.ihf.rwth-aachen.de>
Subject: Re: problem building lang/icon on 2.0.5R
To: asami@cs.berkeley.edu (Satoshi Asami)
Date: Mon, 17 Jul 1995 10:30:02 +0200 (MET DST)
Cc: thomas@ghpc8.ihf.rwth-aachen.de, ports@freebsd.org
In-Reply-To: <199507141225.FAA11512@silvia.HIP.Berkeley.EDU> from "Satoshi Asami" at Jul 14, 95 05:25:15 am
Organization: Institut f. Hochfrequenztechnik, RWTH Aachen
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Content-Length: 1163      
Sender: ports-owner@freebsd.org
Precedence: bulk

Satoshi Asami wrote:
> 
>  * More data:  make runs fine if my login shell is tcsh.  Normally, I
>  * use pdksh and _did_ use pdksh on 2.0 as a login shell when I ported
>  * icon.
> 
> I just tried it on thud, it built fine both with bash and ksh,
> although I didn't change my login shell to either.  I'll try changing
> the shell tomorrow.
> 
> By the way, thud is running (what is belived to be pretty close to)
> 2.0.5R.

I checked again on a machine here at work and later at home.  It has
nothing to do with the shell.  It's just that I have enabled
ENABLE_STARTUP_LOCALE in my .kshrc and set LANG to de_DE.ISO8859-1.
If I unset ENABLE...  icon builds fine.  I have currently no idea
why the locale setting would upset the parser in rtt; the parsed
file (src/runtime/fmisc.r) does not contain characters outside the
ASCII range (some ^L's, but they are also in other .r files), so any
ctype functions should work regardless of the current locale.  No, I
don't know where exactly it hangs.  I chickened out and added a call
to setlocale in more(1) instead.  That's why I had the ENABLE...
set in the first place.  I will mail the diffs to the bugs list.

tg