httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Randy Terbush <ra...@zyzzyva.com>
Subject Re: Finish 1.2!
Date Sat, 11 Jan 1997 16:51:37 GMT
Jim Jagielski sez:
> Marc Slemko wrote:
> > I guess a HAVE_SNPRINTF would be more obvious to find the fix when
> > things broke during compilation...
> > 
> > 
> > No snprintf:
> > 	SunOS falun 5.5 Generic_103093-06 sun4m
> > 	SunOS obed-le0 4.1.4 1 sun4m
> > 	AIX gpu5 1 4 000002929000
> > 	IRIX cab101 5.3 11091812 IP22
> > 	HP-UX nyquist B.10.01 A 9000/735 2014046246 two-user license
> > 
> > snprintf:
> > 	FreeBSD alive.ampr.ab.ca 2.1.5-RELEASE FreeBSD 2.1.5-RELEASE #5: Mon Nov 18 21:48:25
MST 1996     marcs@alive.ampr.ab.ca:/usr/home/marcs/FreeBSD/sys/compile/ALIVE  i386
> > 	BSD/OS valis.worldgate.com 2.0.1 BSDI BSD/OS 2.0.1 Kernel #18: Sun Nov 10 19:11:13
MST 1996 marcs@valis.worldgate.com:/misc/marcs/sys/compile/VALIS  i386
> 
> Looks like most OSs _don't_ have snprintf()... I would guess the best
> solution would be similar to what we do for regex: provide a
> version and allow for people to use that one if they need to or
> want to. No need to rename it since even if the OS has snprintf,
> if Apache is setup to compile it's own, any linker worth it's
> salt will use that version instead of the one in libc (or
> whereever).
> 
> The reason I'm opposed to calling it apache_snprintf() is that it's
> not Apache specific nor "enhanced" for Apache (unless we base it
> on Ben's *printf() stuff). It's simply snprintf().
> 

I would be against our forcing an OS which has snprintf() into using
ours.

As we have discussed in the past, the common way to do this is to
#ifdef NEED_SNPRINTF our suppled snprintf(). Be it a version of Ben's
or whatever, I don't have a preference.






Mime
View raw message