From owner-freebsd-questions@FreeBSD.ORG Fri Jun 13 02:36:30 2003 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9D72F37B401 for ; Fri, 13 Jun 2003 02:36:30 -0700 (PDT) Received: from mx.tele-kom.ru (mx.tele-kom.ru [213.80.148.6]) by mx1.FreeBSD.org (Postfix) with SMTP id 91FF243F75 for ; Fri, 13 Jun 2003 02:36:28 -0700 (PDT) (envelope-from doublef@tele-kom.ru) Received: (qmail 83231 invoked by uid 555); 13 Jun 2003 13:36:25 +0400 Message-ID: <20030613093625.83228.qmail@mx.tele-kom.ru> Received: from (213.80.149.160) by t-k.ru with TeleMail/2 id 1055496984-83219 for doublef@tele-kom.ru; Fri, 13 Jun 13:36:24 2003 +0400 (MSD) Date: Fri, 13 Jun 2003 13:24:18 +0400 (MSD) To: "Mark Miller" From: DoubleF cc: Robert Huff cc: freebsd-questions@freebsd.org Subject: Re: Are there STABLE/CURRENT/RELEASE tags for ports? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 13 Jun 2003 09:36:30 -0000 About CTM vs cvsup: CTM by ftp is faster for my dial-up, so I use it. I don't really mind storing extra 20M on my disk compared to cvs-downing and upping again... About disk space: for me an extra "current" /usr/local takes 3G. It isn't an awful lot, and it may save my ass someday... > Mark Miller writes: > > > On a more pragmatic note, are there any particular reasons that > > port maintainers can't use the -STABLE tag for their updates? It > > seems like a general guideline of "stable lags current by X > > weeks" might help things tremendously. In fact, from this point of view all ports should be considered "stable" except "-devel" ones which are current. It's just that "stable" isn't. If you just like the "-STABLE" tag - that's not a problem;) About "stable lags current by X weeks": what you are asking for is probably another ports tree. There were such statements even here on questions@. Everyone wants to build something rock solid and noone (me included) seems to have an idea of how to do it. If you can tell why your tree will be better, noone will be against... > I've never put together a port, but mu understanding is that > authors have two ways to do this, should they need to: > > set certain variables in the Makefile > > test (IFDEF/IFNDEF) in the code itself > > My experience: _very_ few ports need this; they are written to > not care what version they're on. > > > Robert Huff You might be misunderstanding the question. The update of the OS itself doesn't harm ports (in my humble experience). FreeBSD doesn't break ports, ports break ports (mostly by asynchrounous behavior). I wish I could tell you how to change the structure to remedy the situation - but I have no idea except for what I've already told you. Indeed this should be brought to ports@. DoubleF