From owner-cvs-all  Thu Apr 16 18:24:09 1998
Return-Path: <owner-cvs-all@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id SAA08112
          for cvs-all-outgoing; Thu, 16 Apr 1998 18:24:09 -0700 (PDT)
          (envelope-from owner-cvs-all@FreeBSD.ORG)
Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA07988;
          Fri, 17 Apr 1998 01:23:36 GMT
          (envelope-from asami@vader.cs.berkeley.edu)
Received: from silvia.HIP.Berkeley.EDU (ala-ca34-10.ix.netcom.com [207.93.143.138]) by vader.cs.berkeley.edu (8.8.7/8.7.3) with ESMTP id SAA01787; Thu, 16 Apr 1998 18:23:30 -0700 (PDT)
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.8/8.6.9) id RAA10861; Thu, 16 Apr 1998 17:27:52 -0700 (PDT)
Date: Thu, 16 Apr 1998 17:27:52 -0700 (PDT)
Message-Id: <199804170027.RAA10861@silvia.HIP.Berkeley.EDU>
To: ache@nagual.pp.ru
CC: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG
In-reply-to: <19980417031022.02794@nagual.pp.ru> (message from =?koi8-r?B?4c7E0sXKIP7F0s7P1w==?= on Fri, 17 Apr 1998 03:10:22 +0400)
Subject: Re: cvs commit: src/etc make.conf
From: asami@FreeBSD.ORG (Satoshi Asami)
Sender: owner-cvs-all@FreeBSD.ORG
Precedence: bulk

 * Please add proper description in good English, I am not able to do it.

I can certainly do that (even though my Anglais is not that good), if
we agree what to put there.

 * > (2) The MOTIFLIB description is not appropriate anymore, now that
 * >     MOTIF_STATIC is given as an example here.
 * 
 * It should be changed too, I agree.

I didn't think you disagreed, I was just pointing out that you should
read the stuff around something before adding a new line. :>

 * I think system should be comfortable for developers first and next for
 * ordinary users.

I'm not sure if I agree with that, but that's another point.  However,
do developers really need MOTIF_STATIC?  That makes EVERYTHING link
with static libXm, making that person more of a package builder (one
who builds packages to distribute to other users, not one who builds
package for testing).  I think it's more likely that they need it on a
per-port basis (say, for testing mozilla that doesn't work with
dynamic libXm), in which case the place to fix it is in that port's
Makefile (or Makefile.i386 if you don't want to touch a file
controlled by CVS), not /etc/make.conf.

My fear is that /etc/make.conf is getting too clustered with stuff
that is relevant to only a small set of users, thus making it harder
for people to maintain.  It's already getting a PITA to merge new
changes from /usr/src/etc/make.conf.

Ports developers are supposed to read the handbook as well as the
header of bsd.port.mk, where all these (and other milliards of
variables) are described in great detail.

 * > Maybe it's time for /etc/make.conf.ports?  (No I'm not kidding.)
 * 
 * Yes, good idea.

Seriously though, I don't envision that as a description and default
values for all possible variables.  Those should go in bsd.port.mk,
not only because one thing should be described in as few places as
possible but also because /etc won't be upgraded as part of "make
world".

That means make.conf.ports is empty by default where people can
add things as they wish.  But they can do that to make.conf already,
so I guess it makes the whole thing rather pointless. :<

Satoshi

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message