Hi Matthieu! > Stefan Reichör wrote: >> Hi! >> >> Until the problems with my Windows web hoster are fixed, I have >> created an experimental bzr repository for DVC. >> >> bzr still feels very slow on my 56k Modem dial up line. > > Yes. The knits thing improved the situation a bit, but there are still > progresses to be made for performances in bzr. hg is obviously faster as > of today. > >> So I didn't include the gnu arch history. >> I think we don't need the history in the repository to proceed. >> It is available if needed in our Gnu arch repository. > > That's fine. The only thing is: we can not easily merge from baz to bzr > with this approach. So, we should decide on a precise date for the > actual migration and ask everybody to submit patches only with bzr from > that day.
Since there is not much activity on DVC at the moment, I think we won't have problems with that step. > Another remark : you did a "bzr init; bzr add", I ask other people to do > a "bzr get" or "bzr branch" to get their branch, and not starting the > project from scratch : otherwise, the file identity and the history > won't allow merging. I don't understand what you mean here. * Is it possible to branch from from the GNU Arch archive? * What merge cases do you think of? > Otherwise, the best way is probably to send plain patches by email. > >> The branch contains the revision [EMAIL PROTECTED]/dvc--dev--0--patch-232 >> >> You should get (from an unix web hoster) via: >> bzr get http://xsteve.nit.at/dvc >> >> Could you please try to check out a copy. > > worksforme. So, how should we proceed? My suggestion is to import the project to bzr without history. The drawback is that we loose the ability to merge older patches. But I am not aware of such patches. So that should not be a problem. And if someone has a patch, it can easily extracted as diff and applied. Other problems that I have missed? Stefan. _______________________________________________ Dvc-dev mailing list [email protected] https://mail.gna.org/listinfo/dvc-dev
