On Tue, Sep 9, 2008 at 5:48 PM, Eric Kow <[EMAIL PROTECTED]> wrote: > On Tue, Sep 09, 2008 at 10:06:27 -0400, David Roundy wrote: >> It's already applied and is being pushed, and I see no reason to >> short-circuit the normal testing framework for this sort of thing. > > Oh, I think my definition of fast-track is actually pretty slow because > I also run darcs check (preferably on the fast machine at work). But > I'm happy to stick with the old unstable -> stable route for code patches.
But you probably don't run darcs check on a machine with ghc 6.6, so you'd probably be in danger of letting patches through that would break darcs on my machines. Better to make the code track a slow track that involves two tests (yours and mine), and preferably also a set of buildbot tests. With that in place, we'd have some degree of confidence that stable is not going to be less stable than unstable. David _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
