Date: Mon, 21 Feb 2005 19:16:45 +0100 From: Kirill Ponomarew <krion@voodoo.oberon.net> To: Michael Nottebrock <michaelnottebrock@gmx.net> Cc: freebsd-ports@freebsd.org Subject: Re: devel/pcre and WITH_UTF8 Message-ID: <20050221181645.GH9175@voodoo.oberon.net> In-Reply-To: <200502211744.52024.michaelnottebrock@gmx.net> References: <20050221142951.GA48781@pc5-179.lri.fr> <20050221153615.GE9175@voodoo.oberon.net> <20050221155832.GJ51280@zi025.glhnet.mhn.de> <200502211744.52024.michaelnottebrock@gmx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Feb 21, 2005 at 05:44:47PM +0100, Michael Nottebrock wrote: > > Yes, you're right, since there are exponentially many combinations of > > options, creating slave ports is not the right way. > > We probably need a mechanism [...] > > Don't overengineer. Slave ports pretty much *are* the right way to deal with > this sort of situation - which is that *some* option is too controverse to be > either default off or default on. This does not at all imply that you need to > translate *every* option there is (or could be imagined) into a slave port. It depends on what you want to do. I could split net/gnunet into: gnunet-gdbm, gnunet-mysql, gnunet-tdb, gnunet-bdb3, gnunet-sqlite, gnunet-ipv6, gnunet-guile, hence we get 7 slave ports and one master port. I bet there are people who would think it would be useful to split it into parts, but since all these years I still object to do it. Well, finding another stylish solution instead of slave ports, would be very desirable. -Kirill
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050221181645.GH9175>