From xemacs-m  Fri Dec 13 16:17:59 1996
Received: from venus.Sun.COM (venus.Sun.COM [192.9.25.5]) by xemacs.cs.uiuc.edu (8.8.3/8.8.3) with SMTP id QAA09807 for <xemacs-beta@xemacs.org>; Fri, 13 Dec 1996 16:17:58 -0600 (CST)
Received: from Eng.Sun.COM ([129.146.1.25]) by venus.Sun.COM (SMI-8.6/mail.byaddr) with SMTP id OAA12470; Fri, 13 Dec 1996 14:17:28 -0800
Received: from kindra.eng.sun.com by Eng.Sun.COM (SMI-8.6/SMI-5.3)
	id OAA07199; Fri, 13 Dec 1996 14:17:25 -0800
Received: from xemacs.eng.sun.com by kindra.eng.sun.com (SMI-8.6/SMI-SVR4)
	id OAA15921; Fri, 13 Dec 1996 14:17:24 -0800
Received: by xemacs.eng.sun.com (SMI-8.6/SMI-SVR4)
	id OAA00611; Fri, 13 Dec 1996 14:17:23 -0800
Date: Fri, 13 Dec 1996 14:17:23 -0800
Message-Id: <199612132217.OAA00611@xemacs.eng.sun.com>
From: Martin Buchholz <mrb@Eng.Sun.COM>
To: Jareth Hein <jhod@po.iijnet.or.jp>
Cc: XEmacs Beta Mailing List <xemacs-beta@xemacs.org>
Subject: Configure and IRIX questions
In-Reply-To: <32B14CC8.14520F5E@po.iijnet.or.jp>
References: <32B14CC8.14520F5E@po.iijnet.or.jp>
Reply-To: Martin Buchholz <mrb@Eng.Sun.COM>
Mime-Version: 1.0 (generated by tm-edit 7.94)
Content-Type: text/plain; charset=US-ASCII

>>>>> "Jareth" == Jareth Hein <jhod@po.iijnet.or.jp> writes:

Jareth> In investigating Dan Pilone's reported problem with configure
Jareth> autotragically attempting to make input-method-xlib.c, as well
Jareth> as my own problem with the native sound detection stuff I have
Jareth> a question: is the preprocessor directive '#elif' part of a
Jareth> recognized standard? The cpp for IRIX6.2 seems to lose badly
Jareth> when encountering this, and it is NOT listed in the man pages
Jareth> as a recognized preprocessor directive.  However, it doesn't
Jareth> generate an error either. I should probably elaborate that my
Jareth> development system currently only has SGI's development
Jareth> environment on it.

Jareth> If it is not a standard, shouldn't we eliminate dependance on
Jareth> it? Do any other systems have a similar problem?

It's hard to believe #elif isn't available on all meaningful cpp's
yet.  Even 19.14 has a few #elif's in various Makefiles.  I'm pretty
sure #elif is part of the ANSI standard.

Part of the problem is that we can't use many of the modern tokenizing
cpp's (eg cc -E) because they add disruptive blanks into the
Makefiles.  So we have to use older cpp's, such as Sun's
/usr/ccs/lib/cpp, which doesn't add the blanks, but is modern enough
to have #elif.

Another reason not to use cpp to preprocess our Makefiles.

Martin

