From owner-freebsd-ports  Tue Jan 14 04:35:16 1997
Return-Path: <owner-ports>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.4/8.8.4) id EAA10406
          for ports-outgoing; Tue, 14 Jan 1997 04:35:16 -0800 (PST)
Received: from dfw-ix5.ix.netcom.com (dfw-ix5.ix.netcom.com [206.214.98.5])
          by freefall.freebsd.org (8.8.4/8.8.4) with SMTP id EAA10401
          for <ports@FreeBSD.ORG>; Tue, 14 Jan 1997 04:35:13 -0800 (PST)
Received: from silvia.HIP.Berkeley.EDU (wck-ca21-14.ix.netcom.com [207.94.231.110]) by dfw-ix5.ix.netcom.com (8.6.13/8.6.12) with ESMTP id EAA28114; Tue, 14 Jan 1997 04:34:39 -0800
Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.4/8.6.9) id EAA04509; Tue, 14 Jan 1997 04:34:37 -0800 (PST)
Date: Tue, 14 Jan 1997 04:34:37 -0800 (PST)
Message-Id: <199701141234.EAA04509@silvia.HIP.Berkeley.EDU>
To: eivind@dimaga.com
CC: ports@FreeBSD.ORG
In-reply-to: <3.0.32.19970114131125.00a753f0@dimaga.com> (message from Eivind Eklund on Tue, 14 Jan 1997 13:11:26 +0100)
Subject: Re: AfterStep in -current
From: asami@vader.cs.berkeley.edu (Satoshi Asami)
Sender: owner-ports@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

 * Why is it nescarry to keep track?  I suggest just copying the version
 * number from the version the port author is running to somewhere in the
 * port, and just let that be a minimum.  No changes until the port is
 * 'naturally updated', whereupon people might again be required to update
 * bsd.ports.mk.

Oh, I see.  I guess we can even automate it, just like files/md5
creation.

asami$ make makeid
asami$ cat files/portid
1.248
asami$
<cvs commit, someone else downloads it>
eivind% make
This port requires bsd.port.mk rev. 1.248; you have 1.247.
*** Error code 1

Stop.
eivind% fetch ftp://ftp.freebsd.org/pub/FreeBSD/FreeBSD-current/....

 :)

 * Default to no version requirement.  No need to update old ports.

Actually, if we are going to do something like this, we should make it 
mandatory, to reduce the chance of mistakes.

 * Of course.  The point of this was to make it easier to avoid spurious
 * bug-reports of the type I gave.
 * 
 * Not a pet idea of mine; just something I thought might save _you_ some work.

Thanks, but I still wonder if that kind of case is really that
frequent. ;)

Also, another problem is that if we go to twin-branch development (for
2.2.5 and 3.0, we are planning this after the 2.2R) on the ports tree,
it is quite possible that we'll have two versions of bsd.port.mk.  We
then can't easily compare the revisions on different cvs branches.

But nonetheless, this is a very interesting idea.

Satoshi