On Wed, Nov 29, 2017 at 03:36:07PM -0700, Warren Young wrote: > On Nov 29, 2017, at 3:10 PM, Joerg Sonnenberger <jo...@bec.de> wrote: > > > > On Wed, Nov 29, 2017 at 12:02:11PM -0500, Richard Hipp wrote: > >> On 11/29/17, Jacob MacDonald <jaccar...@gmail.com> wrote: > >>> Ah, I see. It seems strange to me that the policy didn't get set when I > >>> cloned. How can I avoid the same thing in the future and is there any way > >>> to change the artifacts in the local repository to SHA3? > >> > >> There is no good way to change historical artifacts from SHA1 to SHA3. > > > > There could be a loss-less one-time migration of the whole repo > > We discussed that back around March when all of this was happening. The > simple cases are easy and the hard cases are really hard. > > For example, if checkin [1234abcd] has a comment that refers to ticket > [bcdef522] which in turn refers to artifact ID [6543cfe], is the > migration tool expected to chase and re-point all those links when all > the hashes change?
That can be handled by adding optional aliases. Old links will remain valid even though the integrity of the repository depends on the stronger hash. Joerg _______________________________________________ fossil-users mailing list fossil-users@lists.fossil-scm.org http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
