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 1RNkNl-0002QW-BR for openembedded-core@lists.openembedded.org; Tue, 08 Nov 2011 13:02:25 +0100 Received: from elite.brightsigndigital.co.uk ([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 1RNkHl-0007JC-5J for openembedded-core@lists.openembedded.org; Tue, 08 Nov 2011 12:56:13 +0100 From: Phil Blundell To: Patches and discussions about the oe-core layer Date: Tue, 08 Nov 2011 11:56:12 +0000 In-Reply-To: <7fe88fd604911f096a522deafaaeda3a93f5f966.1320750179.git.xiaofeng.yan@windriver.com> References: <7fe88fd604911f096a522deafaaeda3a93f5f966.1320750179.git.xiaofeng.yan@windriver.com> X-Mailer: Evolution 3.0.2- Message-ID: <1320753373.22985.702.camel@phil-desktop> Mime-Version: 1.0 Subject: Re: [PATCH 7/7] pango: Fix bug 1674 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, 08 Nov 2011 12:02:25 -0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit On Tue, 2011-11-08 at 19:16 +0800, Xiaofeng Yan wrote: > - ${@base_contains('DISTRO_FEATURES', 'x11', '--with-x', '--without-x', d)}" > + ${@base_contains('DISTRO_FEATURES', 'x11', '--with-x', '--without-x', d)} \ > + ${@base_contains('DISTRO_FEATURES', 'directfb', '--with-x=no', '--with-x=yes', d)}" That last line looks fairly wrong. If I am understanding this right then someone with neither x11 nor directfb set would get "--without-x --with-x=yes", and someone with directfb but not X would get "--without-x --with-x=no" which surely can't be what's wanted. Did you test this patchset? >+ [ ! -d "/etc/pango" ] && mkdir -p /etc/pango/ Why can't you just ship this folder in the package rather than creating it in the postinst? Also, this should probably be ${sysconfdir} rather than hardcoded. p.