On Tue, Mar 11, 2014 at 6:31 PM, Stephan Beal <[email protected]> wrote:
> TL;dr: it can now generate delta manifests. i still have a lingering bug > where the current leaf is not being calculated/set/updated properly, but > other than that all looks good so far :) (and a rebuild fixes that, meaning > the underlying metadata are in order, i'm just missing some db update > somewhere). > PS, an explanation for those who don't know the difference, but might be interested, between "delta" and "baseline" (non-delta) checkins: A "delta manifest" is a checkin record which is generated a difference of some other checkin (independent of the normal "content heritage" of the checkin). This is a completely different form of delta than the normal "delta compression" Fossil uses on all content, and the basis for a delta checkin is often not its immediate parent. Delta checkins can be much, much smaller than baseline checkins. The checkin linked in the previous mail was something like 850 (delta) bytes vs 13k (baseline) checkin record. Repos with more files will generally show a much more dramatic difference in checkin record size. Interestingly, the main fossil repo doesn't have any deltas because fossil(1) only generates them when it's told to or when the repo has seen at least one delta before (fossil's main repo has not). libfossil, OTOH, defaults to using deltas unless told not to or unless the delta would not be "appreciably smaller" than its baseline (if a delta manifest is "too big" it gets discarded and re-calculated as a baseline manifest). -- ----- stephan beal http://wanderinghorse.net/home/stephan/ http://gplus.to/sgbeal "Freedom is sloppy. But since tyranny's the only guaranteed byproduct of those who insist on a perfect world, freedom will have to do." -- Bigby Wolf
_______________________________________________ fossil-users mailing list [email protected] http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
