From owner-freebsd-hackers  Fri Oct 27 06:08:18 1995
Return-Path: owner-hackers
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id GAA17496
          for hackers-outgoing; Fri, 27 Oct 1995 06:08:18 -0700
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id GAA17487
          ; Fri, 27 Oct 1995 06:08:12 -0700
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id XAA13158; Fri, 27 Oct 1995 23:04:39 +1000
Date: Fri, 27 Oct 1995 23:04:39 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199510271304.XAA13158@godzilla.zeta.org.au>
To: hackers@freefall.freebsd.org, hsu@freefall.freebsd.org
Subject: Re: cpp question
Sender: owner-hackers@FreeBSD.org
Precedence: bulk

>Well, if you use gcc -E or an ANSI cpp, you can use #name.

If you use a 4.4BSD derived system, then you can #include
<sys/cdefs.h> (perhaps indirectly) and use __STRING(), which is
supposed to hide the unportability of #name.  Don't use it in
portable code.  __STRING() and __P(() are less portable than #name
or prototypes.

For portability, you should define your own macros, e.g., for
ANSI:

#define P__(x)		x
#define STR(x)		STR1(x)
#define STR1(x)		#x

STR(x) should normally be used instead of #x even if there are
no portablility considerations, since STR1(x) usually does the
wrong thing if x is a macro (x doesn't get expanded).

The nested macro trick for STR(x) doesn't work in traditional
mode and isn't used in <sys/cdefs.h>.

Bruce