[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: gnats, automake, and GCC_CFLAGS

From: Chad Walstrom
Subject: Re: gnats, automake, and GCC_CFLAGS
Date: Fri, 06 Jan 2012 22:40:13 -0600
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:8.0) Gecko/20111124 Thunderbird/8.0

Copying up to the list, since this pertains to users. Hey! We're back! :) Alex and I have been cleaning up the compilation process a bit, with the code sitting in CVS/HEAD. The conversion to Automake left some flag definitions out of the mix, and changes in autoconf have caused a few breakages.  I'm about to make a commit to HEAD to incorporate some of our changes.  It would be nice to get a 4.2 out the door sooner than later. Any volunteers to help?


On 01/06/2012 01:07 AM, Alex Manoussakis wrote:
I've seen an example of declaring this explicitly before calling the
stdio header:

#declare _POSIX_SOURCE = 1
#include <stdio.h>
I believe this is the right way to go. I've been chewing on this one
since last night, and taking into consideration the following
recommendation by Paul Eggert on 8/1/2000 in the address@hidden list (

"Sorry that my suggestions waffled about _POSIX_SOURCE, but it is a
problematic area.  You can't define _POSIX_SOURCE or _POSIX_C_SOURCE
unconditionally, as that causes possibly-undesirable changes on
Solaris and many other platforms.  On the other hand, if you don't
define thes macros up-front on hosts like Minix, you either need hacks
like AC_MINIX and AC_ISC_POSIX, or you need to have a separate
autoconf macro for fileno, tzset, etc.; and both these alternatives
are more unpleasant than simply defining one of those feature-test
macros if <stdio.h> doesn't define fileno."

I've also noticed that we're putting most of our includes in the gnats.h
header, then including that file everywhere. I'm not really keen on
fixing this right now, but I'm not in favor of this "include everything"
approach.  *shrug*  For now, I'll throw the _POSIX_SOURCE definition
just above stdio.h.

Then maybe we can dive into this unused "input" function.


Chad Walstrom | Technical Support Engineer
p: (651) 757-4136 or (866) 276-5583 ext. 136 f: (651) 665-0943
Customer Support: 800.314.0147 | address@hidden
Sign up for training at
Get knowledge at

reply via email to

[Prev in Thread] Current Thread [Next in Thread]