Date: Sat, 25 Jan 2003 19:42:44 +0300
From: "Eli Zaretskii" <eliz AT is DOT elta DOT co DOT il>
Sender: halo1 AT zahav DOT net DOT il
To: rich AT phekda DOT freeserve DOT co DOT uk
Message-Id: <3405-Sat25Jan2003194243+0200-eliz@is.elta.co.il>
X-Mailer: emacs 21.3.50 (via feedmail 8 I) and Blat ver 1.8.9
CC: djgpp-workers AT delorie DOT com
In-reply-to: <3E319B0C.D99F6005@phekda.freeserve.co.uk> (message from Richard
	Dawe on Fri, 24 Jan 2003 19:59:08 +0000)
Subject: Re: ssize_t: int -> signed long [PATCH]
References: <Pine DOT SUN DOT 3 DOT 91 DOT 1030123081522 DOT 15630C AT is> <3E2FB5F0 DOT B7FD75FC AT phekda DOT freeserve DOT co DOT uk> <1659-Thu23Jan2003203612+0200-eliz AT is DOT elta DOT co DOT il> <3E304571 DOT BFA087FF AT phekda DOT freeserve DOT co DOT uk> <3791-Fri24Jan2003214724+0200-eliz AT is DOT elta DOT co DOT il> <3E319B0C DOT D99F6005 AT phekda DOT freeserve DOT co DOT uk>
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> Date: Fri, 24 Jan 2003 19:59:08 +0000
> From: Richard Dawe <rich AT phekda DOT freeserve DOT co DOT uk>
> > 
> > However, isn't ssize_t mandated by some standard?
> 
> Yes, POSIX or SUS. However, it's not defined in <stdio.h>. <stdio.h> is
> allowed to define anything that <stddef.h> does. But ssize_t is not in
> <stddef.h>. It's defined in <sys/types.h>
> 
> So we would pollute <stdio.h> with ssize_t, if FILE._cnt is an ssize_t. Shall
> I revert FILE._cnt to a long?

I'd go for having ssize_t in stdio.h regardless.  IMHO, a typedef does
not do any harm, unless the program which includes stdio.h defines its
own type under the name ssize_t.  But if others feel strongly against
doing that, I guess we need to go back with FILE._cnt.