Lisandro Dalcin wrote: > On Fri, May 15, 2009 at 4:04 PM, Dag Sverre Seljebotn wrote: >> Stefan Behnel wrote: >>> the large number of major changes in cython-devel induces a number of >>> merge >>> conflicts in cython-unstable. I'm very happy to see all those great >>> improvements all over the place, but could those who implemented them >>> please try to re-integrate them with the unstable branch? >> At least for what I should be doing: Because of time constraints I need to >> keep focus on cython-devel myself; I'll be happy to merge next week after >> the release is done. > > Well, in PETSc (not petsc4py) we have a rather strict rule. Any change > that goes to release-maint is immediately merged to -dev. But we never > introduce new features in release-maint.
Yes, it's a lot easier to adapt changes right after applying them to a branch than to do so a couple weeks later when you happen to try a merge. Plus, the latter increases the chance that others stumble into merge hell for you. >> I don't think what's happened in -devel is a blocker for anything that >> should happen in unstable the next week? I agree that there are no blockers for cython-unstable currently, and I was merely saying that I won't be the one who will do that merge. But I'm also suggesting that those who applied these changes should feel responsible for getting them applied to all three active branches. Stefan _______________________________________________ Cython-dev mailing list [email protected] http://codespeak.net/mailman/listinfo/cython-dev
