From xemacs-m  Fri Sep 19 17:52:57 1997
Received: from mercury.Sun.COM (mercury.Sun.COM [192.9.25.1])
	by xemacs.org (8.8.5/8.8.5) with SMTP id RAA15110;
	Fri, 19 Sep 1997 17:52:56 -0500 (CDT)
Received: from Eng.Sun.COM ([129.146.1.25]) by mercury.Sun.COM (SMI-8.6/mail.byaddr) with SMTP id PAA23514; Fri, 19 Sep 1997 15:52:28 -0700
Received: from kindra.eng.sun.com by Eng.Sun.COM (SMI-8.6/SMI-5.3)
	id PAA15124; Fri, 19 Sep 1997 15:52:27 -0700
Received: from xemacs.eng.sun.com by kindra.eng.sun.com (SMI-8.6/SMI-SVR4)
	id PAA21117; Fri, 19 Sep 1997 15:52:25 -0700
Received: by xemacs.eng.sun.com (SMI-8.6/SMI-SVR4)
	id PAA06390; Fri, 19 Sep 1997 15:52:24 -0700
Date: Fri, 19 Sep 1997 15:52:24 -0700
Message-Id: <199709192252.PAA06390@xemacs.eng.sun.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
From: Martin Buchholz <mrb@Eng.Sun.COM>
To: SL Baur <steve@xemacs.org>
Cc: xemacs-beta@xemacs.org
Subject: Re: (eventual) [success][patch] Bern sparc-sun-solaris2.5.1
In-Reply-To: <m2n2lakqe0.fsf@altair.xemacs.org>
References: <m2sov4pezr.fsf@altair.xemacs.org>
	<ocrafhbpz2n.fsf@ml.com>
	<m23en3zgy9.fsf@altair.xemacs.org>
	<ocr67rypvhf.fsf@ml.com>
	<m2n2lakqe0.fsf@altair.xemacs.org>
X-Mailer: VM 6.33 under 20.3 "Vienna" XEmacs  Lucid (beta14)
Reply-To: Martin Buchholz <mrb@Eng.Sun.COM>

>>>>> "sb" == SL Baur <steve@xemacs.org> writes:

sb> Colin Rafferty <craffert@ml.com> writes:
>> I believe that any call to an emacs in the build process should be to
>> the just-built version.  Anything else is inconsistant and error-prone.

sb> Um, I forgot I fixed man/tm/Makefile to not blow chunks on the Japanese
sb> files when presented with XEmacs/Latin-1. :-/

sb> Does this do what you want?

sb> 1997-09-18  SL Baur  <steve@altair.xemacs.org>

sb> 	* Makefile (EMACS): Refer to xemacs binary in source tree.
sb> 	* tm/Makefile (EMACS): Ditto.

sb> Index: man/Makefile
sb> ===================================================================
sb> RCS file: /usr/local/xemacs/xemacs-20.0/man/Makefile,v
sb> retrieving revision 1.13
sb> diff -u -r1.13 Makefile
sb> --- Makefile	1997/08/01 03:28:18	1.13
sb> +++ Makefile	1997/09/19 01:58:44
sb> @@ -23,7 +23,7 @@
sb>  # inherited from the environment.
sb>  SHELL = /bin/sh
 
sb> -EMACS = xemacs
sb> +EMACS = ../src/xemacs
sb>  EMACSFLAGS = -batch -q -no-site-file 
 

EMACS = ../src/xemacs
isn't going to work if using --srcdir=...

Unfortunately, this can't be done truly correctly without having
builddir/man be a separate directory from srcdir/man, in the general
case (like src).

I recommend we half-fix this as follows:

Make sure that when you do
make info
in blddir, it runs
cd man && $(RECURSIVE_MAKE) EMACS=$(blddir)/src/xemacs

so that make info in blddir works, although make info in blddir/man
might not.

The primary maintainer can build the info files the current way
(having to be careful about having xemacs on the path be a --with-mule 
one).  

Having building of info files be slightly hostile is OK, since most
XEmacs site maintainers will not have to do this - XEmacs will come
with pre-built info files (or at least they will be available).
However, we should make sure that if xemacs is latin-1, then all the
info files are built except for the ones actually requiring a mule
xemacs.

Martin

