On Tuesday 2012-06-19 14:15, Richard Guenther wrote:
Just inquiring. If your scripts call `osc rebuild` one time once they see a failed state of some kind, I'd still call that manual. (Because your scripts likely don't run on all projects marked with rebuild=local - unless they are active by default in build-service-2.x)
I don't see why "manual" is bad. In fact we should be perfectly aware about an ABI/API change _before_ we check things in. Being ignorant and relying on automatic rebuilds "fixing" things is even worse than a concious "manual" rebuild.
Some SONAME bumps are only no-brainer changes. Like, adding a member to a struct without change of semantics for existing code. In that case, you *need* a rebuild because of the changed ABI, but that's about it. If the API had changed in that instant as well, I'd see it when the automatically-triggered rebuild failed and/or spewed out new warnings. An example I can remember is when libcryptsetup1 moved to libcryptsetup4. There were some old functions removed which most programs don't seem to have recently used anyway. (And if they did, I'd love to see the failure in the automatically triggered rebuild about symbols not found rather than having a green package that does not install nor run.) -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org