On Fri, Jan 29, 2010 at 09:21:38AM +0100, Jens Seidel wrote: > On Thu, Jan 28, 2010 at 04:22:32PM -0700, Chris Larson wrote: > > On Thu, Jan 28, 2010 at 3:59 PM, Tom Rini <[email protected]> wrote: > > > On Thu, 2010-01-28 at 12:00 -0800, Khem Raj wrote: > > > > Attached is a small hook for updates that are pushed into repo. > > > > > > Is there any way to force the commit to happen, even if the script > > > doesn't like it? Can the script edit the commit message, even? ie if > > > there's no git commit --ignore-prehook type option, could it see if the > > > first line is FORCE or something, edit that out and commit? This will > > > catch the poorly formed commit messages without being a big burden on > > > people doing things that don't fit well into the "module: summary" model > > > (that said, lib*-perl-native: Convert to BBCLASSEXTEND as a first line > > > makes sense to me). > > > > My worry with editing a commit at push time is, the user's local repository > > now no longer matches upstream. > > Right, what about two repositories, one with a hook which denies some > commits and another one without such a hook. The second repository could be > synced with the first via a cronjob and users are always forced to commit > to the first repository except they know *exactly* what they are doing > (maybe by restricting access to a few people only) and want to bypass these > checks?
This looks like a bit overcomplicated sollution. I think that every commit message "on edge" of commit policy can be made to go through. ie: fso_recipes: switch from SRCREV to SRCPV instead of fso recipes: switch from SRCREV to SRCPV lib*-perl-native: Convert to BBCLASSEXTEND could go through as long as it's splited by ' ' is almost as readable as original and easier to s/ /_/ then pusing to different repo and waiting for cron job to sync it, hoping that nobody else didn't commit conflicting change to right repo with right commit message first. Regards, -- uin:136542059 jid:[email protected] Jansa Martin sip:[email protected] JaMa _______________________________________________ Openembedded-devel mailing list [email protected] http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
