From owner-freebsd-ports@FreeBSD.ORG Sat Mar 11 17:44:33 2006 Return-Path: X-Original-To: freebsd-ports@freebsd.org Delivered-To: freebsd-ports@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id ACBDD16A41F for ; Sat, 11 Mar 2006 17:44:33 +0000 (GMT) (envelope-from linimon@lonesome.com) Received: from mail.soaustin.net (mail.soaustin.net [207.200.4.66]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6CF6F43D45 for ; Sat, 11 Mar 2006 17:44:33 +0000 (GMT) (envelope-from linimon@lonesome.com) Received: by mail.soaustin.net (Postfix, from userid 502) id 2E517407C; Sat, 11 Mar 2006 11:44:33 -0600 (CST) Date: Sat, 11 Mar 2006 11:44:33 -0600 To: Aluminium Oxide Message-ID: <20060311174433.GB10679@soaustin.net> References: <1142054372.14978.256358031@webmail.messagingengine.com> <20060311060002.GA40638@xor.obsecurity.org> <1142079054.9555.256373197@webmail.messagingengine.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1142079054.9555.256373197@webmail.messagingengine.com> User-Agent: Mutt/1.5.9i From: linimon@lonesome.com (Mark Linimon) Cc: freebsd-ports@freebsd.org, Kris Kennaway Subject: Re: portsupgrade cvsup refuse file aware? X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 11 Mar 2006 17:44:33 -0000 On Sat, Mar 11, 2006 at 10:40:54PM +1030, Aluminium Oxide wrote: > > > How much better would it be, if pkgdb had a flag to permit the parsing > > > of the refuse file, and honouring the ports tree resulting from cvsup as > > > produced by it? > > > > > > Can we do this? > > > >Not really, this has been discussed many times. > > > > Kris > > Why not? Executive Summary: partial ports trees don't work. Longer Summary: until and unless we (portmgr) wants to enforce a new rule that limits inter-category ports dependencies, there is no way to know which category is a "leaf category" and thus may be safely trimmed, and thus will be able to make a consistent INDEX. Trying to deal with people who have failed to create INDEX is an ongoing, frustrating, and ultimately useless task, and it's much easier just to tell everybody to have complete ports trees. The last *three* times I have asked for a show of hands, "who wants a new enforceable portmgr policy to restrict dependencies in certain categories", I have received a deafening silence (no replies at all). Yes, this topic has been discussed at least three times. As kris hinted, see the archives. mcl