From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail.pbcl.net ([88.198.119.4] helo=hetzner.pbcl.net) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1QgZ6D-0001xV-JH for openembedded-core@lists.openembedded.org; Tue, 12 Jul 2011 11:17:49 +0200 Received: from cambridge.roku.com ([81.142.160.137] helo=[172.30.1.145]) by hetzner.pbcl.net with esmtpsa (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.72) (envelope-from ) id 1QgZ2Q-0002x4-Sw for openembedded-core@lists.openembedded.org; Tue, 12 Jul 2011 11:13:55 +0200 From: Phil Blundell To: Patches and discussions about the oe-core layer Date: Tue, 12 Jul 2011 10:13:54 +0100 In-Reply-To: <1310461868.6337.179.camel@phil-desktop> References: <1122cff5b635c23b36bc76ad58a088f522ac21a5.1310425837.git.sgw@linux.intel.com> <1310461868.6337.179.camel@phil-desktop> Organization: Phil Blundell Consulting Ltd X-Mailer: Evolution 3.0.2- Message-ID: <1310462034.6337.181.camel@phil-desktop> Mime-Version: 1.0 Subject: Re: [PATCH 03/26] shared_lib.bbclass: create common class X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list Reply-To: Patches and discussions about the oe-core layer List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Jul 2011 09:17:49 -0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit On Tue, 2011-07-12 at 10:11 +0100, Phil Blundell wrote: > On Mon, 2011-07-11 at 16:13 -0700, Saul Wold wrote: > > This class is for packages that do not work with debian library naming > > I tend to feel that most of those packages should be fixed so that they > do. If they aren't currently getting renamed then it usually means that > they include too much (i.e. both binaries and libraries, or multiple > libraries) and should be split up so that this doesn't happen. Actually, looking at the patch more carefully I guess this does basically address the first case above (binaries and libraries). But it looks like it will result in the slightly confusing situation where the libraries go into lib${PN} but the corresponding headers are in ${PN}-dev. p.