Hi, On Thu, Aug 19, 2010 at 04:14:28PM +0100, Eric Kow wrote: > 1. What kind of indefinite support should we provide for old-fashioned > repositories?
I think doing a get from old fashioned to current format should be always supported. > 2. What pre-requisites need to be fulfilled before we can withdraw > support for old-fashioned repositories? Note that these aren't > pre-requisites for agreeing on the withdrawal; they are things we > need to work towards if we should agree in principle. There shouldn't be many. We should be sure that there is a reasonable migration path. We should also make it possible to write out a non-hashed copy of a repository, although I'd say that using earlier 2.x darcs for that is quite acceptable. Going from old-fashioned to hashed should be possible with all future 2.x versions, I suppose... > 3. What kind of schedule should we adopt in phasing out support for > old-fashioned repositories? > Perhaps we could agree on the principle that darcs proper should always > be able to get an old-fashioned repository (upgrading it as need be?). > I say this because old repositories die hard, as an extreme case, this > ancient repository from the pre darcs-1.0.4, which resurfaced around > 2009-03 [1]. Agreed. > 2. Phase-out pre-requisites? > ---------------------------- > Would it be reasonable to want the large majority of repositories to be > hashed before we even start putting out deprecation warnings? [or is > this one of those situations where people are only going to move if we > force them to?]. I don't think waiting is going to help much. It'll take maybe another 2 years before actual darcs release with minimal old-fashioned support is widely adopted. > How about ensuring that for users who are still deliberately using > hashed repositories, hashed repositories are no longer a significant > performance regression? It would be easier to address this *after* removing old-fashioned support. > How about answering concerns Ian may have about hashed repository > performance in practice? > > - Garbage collection? A bug. Again, something that will be easier to fix with cleaner code. > - Large _darcs/pristine.hashed due to insufficient bucketing? On this, we are probably stuck due to backwards compatibility within hashed. I'll see what can be done. Once again, we first need to clean up the code to know better. > Petr suggests a sort of withering-away approach... maybe there's a more > deliberate strategy? I am all in favour of going through the code and killing old-fashioned deliberately. I have actually started doing just this. Yours, Petr. _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
