From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnout Vandecappelle Date: Tue, 18 Jun 2013 08:21:17 +0200 Subject: [Buildroot] Changes in the Buildroot autobuilders In-Reply-To: <20130617175407.GA3202@free.fr> References: <20130616180054.58f52914@skate> <20130616170305.GI3495@free.fr> <20130616191806.7b4a899a@skate> <20130616172508.GL3495@free.fr> <20130616200540.5a370504@skate> <20130617074535.GB16699@lukather> <20130617094958.33eaff88@skate> <20130617175407.GA3202@free.fr> Message-ID: <51BFFC5D.4040600@mind.be> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On 17/06/13 19:54, Yann E. MORIN wrote: > On 2013-06-17 11:20 -0500, Spenser Gilliland spake thusly: >> >Are we sure we want to add more variables to the .mk files. One of >> >Buildroot's main differentiators is the cleanliness of these files in >> >comparison to other packaging methods. +1 At least on the short term, the number of maintainers will be small and the number of packages per maintainer will be large. Some maintainers will also be fairly ephemeral - e.g. Maxime added support for systemd a year ago, but probably by now he doesn't consider himself a maintainer of it anymore. Also the rules may be a little more complex than based on package name. So I think a nice solution would be to add a script in support/scripts that explicitly encodes all the maintainers - maintainers can add themselves with patches (just like in the kernel). The a.b.o would fetch the script once a day just before sending out its daily mails, and call it with everything it knows about the build failure set in the environment. >> > >> >As an alternative, this information could be pulled from the git log >> >and anyone who has made a change to the package directory in the past >> >would be subscribed to errors. > No, I am not in favour of this solution. > > Being a 'maintainer' should be voluntary, opt-in, and explicit. +1 too. However, Spenser points out in another mail: > I think many people would like to know if something they did may have > triggered a bug, not just the maintainer of the package. I could agree with sending mails automatically to the authors of a package commit if the commit was done in the last week or so. Regards, Arnout -- Arnout Vandecappelle arnout at mind be Senior Embedded Software Architect +32-16-286500 Essensium/Mind http://www.mind.be G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle GPG fingerprint: 7CB5 E4CC 6C2E EFD4 6E3D A754 F963 ECAB 2450 2F1F