Date: Wed, 4 Feb 1998 19:40:33 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: Vik Heyndrickx <Vik DOT Heyndrickx AT rug DOT ac DOT be>
cc: DJ Delorie <dj AT delorie DOT com>, djgpp-workers AT delorie DOT com
Subject: Re: char != unsigned char... sometimes, sigh
In-Reply-To: <34D72299.3037@rug.ac.be>
Message-ID: <Pine.SUN.3.91.980204193752.26711B-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Tue, 3 Feb 1998, Vik Heyndrickx wrote:

> I cannot imagine that most gnu utilities for example should have to cope
> with this problem, because they have to run on both kind of
> architectures.

I have seen several GNU packages that test at configure time whether char
is signed or unsigned.  Apparently, their way of running on both kinds of 
architecture is by changing some of the code on each case.

> I even cannot imagine why configure should test this since using
> 'unsigned char' and 'signed char' would take away this problem in the
> first place.

One reason is because some pre-ANSI compilers don't accept "signed char".