Mailing-List: contact new-httpd-help@apache.org; run by ezmlm
Precedence: bulk
Reply-To: new-httpd@apache.org
From: Jim Jagielski <jim@jaguNET.com>
Message-Id: <200101090122.UAA04514@devsys.jaguNET.com>
Subject: Re: cvs commit: httpd-2.0/server config.c
To: new-httpd@apache.org
Date: Mon, 8 Jan 2001 20:22:00 -0500 (EST)
Reply-To: jim@jaguNET.com
In-Reply-To: <3A5A5528.7101617A@algroup.co.uk> from "Ben Laurie" at Jan 09,
 2001 12:02:48 AM
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

Ben Laurie wrote:
> 
> > Are you saying that when autoconf generates C code to test for a given
> > feature, it almost always fails when -Werror is used?
> 
> Correct!
> 
> > If so, that is completely bogus, and it is just one more reason that these
> > projects aren't as good as they could/should be.  :-(
> > 
> > Autoconf should just ignore EXTRA_CFLAGS however, so what went wrong when
> > using EXTRA_CFLAGS to set -Werror?
> 
> It didn't ignore them! Either that, or I'm losing my mind... oh, look,
> I'm not, but configure is:
> 

configure doesn't ignore EXTRA_CFLAGS. At least not the way
that we've coded it, because we wanted to maintain the
1.3 use of the EXTRA_* family. EXTRA_CFLAGS is used to
set extra settings to CFLAGS required for whatever.
Same would have happened if EXTRA_* didn't exist, since
what we would have done is CFLAGS="..." ./configure and
configure would *still* use -Werror. It's because
autoconf uses the entire CFLAGS stuff when it tries to
compile it's test programs, as well it should, since
the existance of functions, etc could depend on those
(eg: -DPOSIX).

-- 
===========================================================================
   Jim Jagielski   [|]   jim@jaguNET.com   [|]   http://www.jaguNET.com/
          "Casanova will have many weapons; To beat him you will
              have to have more than forks and flatulence."