On Thu, Feb 20, 2014 at 02:17:52PM -0800, J. Liles wrote: > On Thu, Feb 20, 2014 at 2:01 PM, Lieven Moors <[email protected]> wrote: > > > On Mon, Feb 17, 2014 at 09:43:38AM -0800, J. Liles wrote: > > > If a 'snapshot' file exists and is newer than the 'history' file, then > > the > > > Compaction operation is equivalent to replacing the contents of latter > > with > > > the former. > > > > > > If the 'snapshot' file doesn't exist or the 'history' file is newer than > > > it, the 'snapshot' file can be brought up to date by simply loading the > > > project in Non Timeline and quitting normally. > > > > > > > Does that also mean that the snapshot file is normally updated whenever > > you quit a non-timeline session? Then compaction would mostly just > > delete history? > > > > Yes. The only reason the snapshot would be out of sync with the history is > if non-timeline is closed abnormally. If this happens, upon the next > opening, non-timeline will load by replaying the entire history instead of > utilizing the outdated snapshot. You can detect this scenario in your > scripts by comparing the file timestamps. > > > > > > I'm actually trying to avoid loading the sessions in non-timeline, > > because I want to run through all git commits in repository, and only > > keep those audio files that are referenced by the timeline sessions. > > So it should be scriptable... > > > > Wait... Am I to understand that you're storing the actual audio files as > objects in git?
No, although I've tested that scenario for a while with git-annex. Git-annex worked well for me, although it puts an extra component in the mix, and I try to keep it as simple as possible for now.
