X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-workers-bounces using -f
Date: Wed, 20 Feb 2002 09:43:05 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Tim Van Holder <tim DOT van DOT holder AT pandora DOT be>
cc: djgpp-workers AT delorie DOT com,
   Juan Manuel Guerrero <ST001906 AT HRZ1 DOT HRZ DOT TU-Darmstadt DOT De>
Subject: Re: bison and djgpp.env
In-Reply-To: <1014190319.14302.6.camel@bender.falconsoft.be>
Message-ID: <Pine.SUN.3.91.1020220093921.21448M-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
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


On 20 Feb 2002, Tim Van Holder wrote:

> I think this has been discussed before, but I couldn't find it in the
> archives: do we still want to support really old bison versions (i.e.
> those that need BISON_HAIRY and BISON_SIMPLE to be set) in 2.04?

I think we do want to support them.  People shouldn't be required to 
update all of their binaries together with djdev.

What harm does it make to continue having these variables?  Juan was 
talking about a test suite, and adding the `+' solves that.  I think 
that's good enough, and we shouldn't be unreasonably radical.

Also note that the v2.03 refresh will have those changes, so it's not 
only v2.04 that we need to consider.

> I believe 2.04 should ditch the variables altogether (and bison's README
> should probably tell users to remove old skeletons from $DJDIR/lib to
> avoid problems).

Why does it make sense to break back compatibility, unless we have a very 
good reason?