From owner-svn-ports-all@FreeBSD.ORG Sun Sep 2 18:37:54 2012 Return-Path: Delivered-To: svn-ports-all@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 1033) id 332301065674; Sun, 2 Sep 2012 18:37:54 +0000 (UTC) Date: Sun, 2 Sep 2012 18:37:54 +0000 From: Alexey Dokuchaev To: Alberto Villa Message-ID: <20120902183754.GI10884@FreeBSD.org> References: <201209010946.q819kRxc085704@svn.freebsd.org> <20120902174533.GE10884@FreeBSD.org> Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.1i Cc: svn-ports-head@freebsd.org, svn-ports-all@freebsd.org, ports-committers@freebsd.org Subject: Re: svn commit: r303465 - head/multimedia/mlt X-BeenThere: svn-ports-all@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: SVN commit messages for the ports tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 02 Sep 2012 18:37:54 -0000 On Sun, Sep 02, 2012 at 08:11:47PM +0200, Alberto Villa wrote: > On Sun, Sep 2, 2012 at 7:45 PM, Alexey Dokuchaev wrote: > > It seems that we commonly do not put space after GTK+ (e.g. GTK+1, GTK+2). > > And that's wrong, like for Qt 4 or KDE 4. That's why I fixed it in > bsd.options.desc.mk. I'm open to discussion on this. Hmm, now that you've said it, I start to hesitate myself. I hereby withdraw my objection, as long as everything (GTK+, Qt, KDE) stays consistent. > > particular ABI is indeed required (in this case, it should been mentioned > > in commit log or Makefile comment). > > I always find this an interesting point, but never managed to read > anything near to "official" about it, so I always left it as it was at > the beginning. Is there some thread discussing it? True, the whole issue is pretty moot. I, for instance, not so long ago was supporting ABI versions, until I was pointed out that it brings more trouble than benefit in most of the times. I will spare you the details, but the most important issue was that: in 99% cases requesting particular ABI version is superfluous (or ever wrong), and that is just complicating ports upgrades (like if you have libfoo.1 installed and perfectly happy with it, but some port now wants libfoo.2 which forces you to upgrade libfoo, even that the software itself does not really care). As a nasty side effect, it complicates binary packages proper handling. (Like why Fedora's packages are so easy to keep rolling on, why ours binary updates result in tons of warnings.) > Thank for reviewing (really). ;) No, thanks *you* for accepting it constructively; I really see how we had improved since two-three years ago (then people used to say "NOT RELEVANT FOR THE BUILD!!11" and subsequently ignored any reviews/critics). ./danfe