From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roel Janssen Subject: Re: [PATCH] import: Update Bioconductor release to 3.5. Date: Fri, 28 Apr 2017 11:11:12 +0200 Message-ID: References: <87inlqjqbj.fsf@elephly.net> <87efwethu9.fsf@gnu.org> <87pofyat7b.fsf@elephly.net> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:49729) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d41vv-0001D1-C8 for guix-devel@gnu.org; Fri, 28 Apr 2017 05:11:24 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d41vs-0000Zh-3I for guix-devel@gnu.org; Fri, 28 Apr 2017 05:11:23 -0400 In-reply-to: <87pofyat7b.fsf@elephly.net> List-Id: "Development of GNU Guix and the GNU System distribution." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: guix-devel-bounces+gcggd-guix-devel=m.gmane.org@gnu.org Sender: "Guix-devel" To: Ricardo Wurmus Cc: guix-devel Ricardo Wurmus writes: > Roel Janssen writes: > >>> I’d be happy if you could take care of the mass update. I should note >>> that sometimes new inputs are required. To find them I usually run the >>> update in a separate branch where I’ve applied changes to import anew >>> and compare with the existing package expression when updating. It’s on >>> my list to clean this all up and submit my changes for review. It’s >>> ugly but it works pretty well. If you’re interested I could send you a >>> patch. >> >> How do you test which inputs are required? I built all the bioconductor >> packages and fixed their builds by adding inputs wherever that was >> needed. This does not take care of inputs that are no longer needed (if >> any..). Any way I can test that in a convenient way? > > Attached is my rough patch set. Apply this and run the update with > “guix refresh -t bioconductor,cran -u”. It will tell you to “consider” > removing or adding inputs. Some of the suggestions are wrong, but it > really means well :) > > It’s terrible code in some places. Don’t look too closely. Thanks! I think I've processed the mass-update now. It looks like we have only gained dependencies. (Not a lot though). I kept the SVN repository URL for Bioconductor for now, instead of the Github mirror. AFAIU, Bioconductor is going to transition to Git (and Github) before the next release, which deprecates the 'Bioconductor-mirror' repository. Since package maintainers get the heads-up and probably have to take some action when this transition occurs, I will make sure we transition without service disruption. :) Kind regards, Roel Janssen