I think I just recompiled xorg for a few small config file changes... without any compiler option changes or useflag changes... I would have been more than happy to spend 10 minutes typing in config file changes, rather than wait 4 hours for xorg to compile... This would allow people to feel freer in customizing the software... Like, I could have a version of the config file that I use in portage under my special name, and fred could, and so could joe.. and we'd be able to switch easily. Additional possible features include, a group of these config file ebuilds customized for certain purposes (roles)... So i could say, emerge Dedicated_Wiki_server, and have all the config files show up right for all the ebuilds. Basically I'm starting to get ticked off at recompiling everthing for the minorest little typo, and I think there are other bugs out there that are related to this, but I think that this is a logical first step Reproducible: Always Steps to Reproduce: 1. 2. 3.
Care to share which xorg you're talking about?
Well, my understanding of the policy is that these kinds of changes do not warrant new revisions. The actual judgement is not that easy, and also depends on the size of the package, and the impact of the change. In any case most revisions are not really necessary for operation. To take that into account does however involve some manual revision of the updates to be performed. I agree that a tool that would allow you to "bump" revisions (like inject used to be useable for) would be helpful for this (you can do it manually). In the end however it might be better to only do updates on security issues (take a look at glsa-check) and improvements to packages you care for. That way you don't get all those recompiles at the cost of not being super-current.
This is not a portage bug or feature request from what I can tell.
x11 herd, please handle this.
In xorg's specific case, I'm rather confident that your thought is not correct on just config file changes in an update. In the more general case, you'll be better off posting to gentoo-dev rather than filing a bug, so I suggest you do that.
