This is the mail archive of the
xconq7@sources.redhat.com
mailing list for the Xconq project.
C99 and Xconq (was Re: Fixes for string functions)
- From: Eric McDonald <mcdonald at phy dot cmich dot edu>
- To: Jim Kingdon <kingdon at panix dot com>
- Cc: sigra at home dot se, <xconq7 at sources dot redhat dot com>
- Date: Thu, 28 Aug 2003 20:44:17 -0400 (EDT)
- Subject: C99 and Xconq (was Re: Fixes for string functions)
On Thu, 28 Aug 2003, Jim Kingdon wrote:
> vsnprintf is in C99. See
> http://www.opengroup.org/onlinepubs/007904975/functions/vfprintf.html
> for a standard.
Maybe we could officially allow C99 into Xconq. It has already
found its way in via C++/C99 double-slash comments. See the
commented out K&R function declarations for example of their
current usage in Xconq (in util.c). (I find it somewhat amusing:
K&R commented out with C++ double-slashes.)
Aside from having vsnprintf and kin available, we would also be
able to use the "inline" keyword for directing the compiler to do
explicit inlining (provided optimization is turned on).
> But you are right, it isn't in C89. At least so far, I've avoided
> assuming it for xconq purposes.
Maybe we should reconsider. We just need to try and verify that
any reasonably modern C compiler/libraries on any of our target
platforms support C99 completely.
I think a reasonable set of compilers to look at would be Visual
C++, Codewarrior, gcc/glibc on Linux and Cygwin, gcc and Sun's C
library on Solaris (if that is currently a workable combination),
and the SunWorkshop C compiler. I left out *BSD, HPUX, Irix,
OSF/Tru64, and AIX because I haven't seen any indication that
anyone is running Xconq on those platforms.
Any thoughts?
-Eric