Mailing-List: contact new-httpd-help@apache.org; run by ezmlm
Precedence: bulk
Reply-To: new-httpd@apache.org
Date: Fri, 28 Apr 2000 14:33:00 +0200
From: "Ralf S. Engelschall" <rse@engelschall.com>
To: new-httpd@apache.org
Subject: Re: cvs commit: apache-2.0/src/os/win32/installer/installdll/test
 test.c
Message-ID: <20000428143300.A23117@engelschall.com>
Reply-To: rse@engelschall.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.1.12i
Organization: Engelschall, Germany.


In article <20000428064956.75303.qmail@locus.apache.org> you wrote:
> dougm       00/04/27 23:49:55
> 
>   Modified:    src/lib/apr/file_io/beos readwrite.c
>                src/lib/apr/file_io/os2 dir.c open.c pipe.c readwrite.c
>                src/lib/apr/file_io/unix readwrite.c
>                src/lib/apr/file_io/win32 filedup.c fileio.h filestat.c
>                         pipe.c readwrite.c
>                src/lib/apr/include apr_general.h
>                src/lib/apr/lib apr_pools.c apr_snprintf.c
>                src/lib/apr/locks/os2 locks.c
>                src/lib/apr/locks/win32 locks.c
>                src/lib/apr/misc/win32 names.c start.c
>                src/lib/apr/network_io/os2 sockets.c
>                src/lib/apr/network_io/win32 sendrecv.c
>                src/lib/apr/shmem/unix/mm mm.h mm_alloc.c mm_core.c
>                         mm_global.c
> [...]

Are you sure APR's MM should be converted, too? AFAIK MM inside APR
doesn't include apr_general.h. MM is a stand-alone library which is just
staying inside the APR tree. But it doesn't use APR, so it cannot (and
IMHO also should not) use APR's APR_FALSE/APR_TRUE values.

BTW, I'm also personally not convinced that own APR_{TRUE,FALSE} values
are very useful. FALSE is _ALWAYS_ 0 and TRUE is just not FALSE. So it
is usually never a problem to use

#ifndef FALSE
#define FALSE 0
#endif
#ifndef TRUE
#define TRUE !FALSE
#endif

even if a vendor header or a third-party header already defined FALSE
(if it defined FALSE, it has to be defined to the value 0 and direct
value comparisons with TRUE are de-facto illegal at all). So, I've never
seen a problem with this in practice. NULL, FALSE and TRUE are more or
less de-facto values for C programming and one can count on them. The
reason why one defines it in own headers is just to make sure they are
really always available (in case <stdlib.h> doesn't define NULL, etc).

                                       Ralf S. Engelschall
                                       rse@engelschall.com
                                       www.engelschall.com