From xemacs-m  Fri Jan 10 00:40:17 1997
Received: from venus.Sun.COM (venus.Sun.COM [192.9.25.5])
          by xemacs.cs.uiuc.edu (8.8.4/8.8.4) with SMTP
	  id AAA22447 for <xemacs-beta@xemacs.org>; Fri, 10 Jan 1997 00:40:16 -0600 (CST)
Received: from Eng.Sun.COM ([129.146.1.25]) by venus.Sun.COM (SMI-8.6/mail.byaddr) with SMTP id WAA20066; Thu, 9 Jan 1997 22:39:47 -0800
Received: from kindra.eng.sun.com by Eng.Sun.COM (SMI-8.6/SMI-5.3)
	id WAA28042; Thu, 9 Jan 1997 22:39:44 -0800
Received: from xemacs.eng.sun.com by kindra.eng.sun.com (SMI-8.6/SMI-SVR4)
	id WAA12004; Thu, 9 Jan 1997 22:39:42 -0800
Received: by xemacs.eng.sun.com (SMI-8.6/SMI-SVR4)
	id WAA24870; Thu, 9 Jan 1997 22:39:41 -0800
Date: Thu, 9 Jan 1997 22:39:41 -0800
Message-Id: <199701100639.WAA24870@xemacs.eng.sun.com>
From: Martin Buchholz <mrb@Eng.Sun.COM>
To: Steven L Baur <steve@miranova.com>
Cc: xemacs-beta@xemacs.org
Subject: Re: 20.0-b34 - no run-in-place??
In-Reply-To: <m27mlm6vrj.fsf@altair.xemacs.org>
References: <199701050651.BAA08127@spacely.icd.teradyne.com>
	<m2wwtsziib.fsf@altair.xemacs.org>
	<199701100313.TAA05464@themom.eng.sun.com>
	<m27mlm6vrj.fsf@altair.xemacs.org>
Reply-To: Martin Buchholz <mrb@Eng.Sun.COM>
Mime-Version: 1.0 (generated by tm-edit 7.100)
Content-Type: text/plain; charset=US-ASCII

>>>>> "sb" == Steven L Baur <steve@miranova.com> writes:

>> The login shell problem is *not* enough of a reason to keep
>> --run-in-place.

sb> Vinnie pointed out in email to me that --run-in-place also allows one
sb> to install binaries to a bin directory in the local source tree.
sb> Without --run-in-place one is at the mercy of hand installed binaries,
sb> or binaries installed from previous versions (look what kind of
sb> confusion that has already generated).  Or one is required is do a
sb> full copying install before running a new beta.  I don't have enough
sb> disk space for that, and I don't expect anyone else to either.

I use --srcdir and just use the generated binaries where they are
created.  My workspaces usually have anywhere from 2 to 5 binaries
associated with them.  If binaries are installed into the source tree
somewhere, shouldn't they be able to find the lisp directories by
snooping around in the directory structure?

Also, the generated xemacs could remember where the srcdir was at
build time, and use that as a `last resort' at run time.  That would
help with your login shell too, where invocation information is useless.

sb> Please reconsider about putting it back.  I am very much opposed to its
sb> removal.

I really thought I was doing the world a favour.  The --run-in-place
seems to me a symptom of one of the biggest bugs with the build
process with most GNU utilities, one that we have avoided - being tied
to a particular install directory.  

Martin

