Hi Peter, Le dimanche, 8 décembre 2013, 03.07:01 peter green a écrit : > cups-filters was getting pretty close to migrating to testing when > 1.0.42-1 was uploaded resetting the counter and failing to build.
Indeed; sorry for this, I didn't expect the new cups-filters version to FTBFS on kFreeBSD. > unless there is a good reason not to* it's generally best to let one > version migrate before uploading another. Especially as cups currently > has broken build-depends in testing** that would have been fixed by > the migration. > > * I apologise if there was a good reason to think that 1.0.41-2 should > not have migrated, I couldn't see any major bugs closed in the > changelog but I may have missed something. The only reason that motivated me to upload 1.0.42 was to stay in close sync with upstream; I have in the past let upstream fly to far ahead (in this and other packages) and it imposed me too much work to get back in sync; I prefer to stay timely close to upstream versions when possible. Second, as we're still outside the freeze (also far from the freeze date), I don't consider the "1.0.42-1 didn't migrate to testing" a big problem. I will make sure to get this fixed and migrated to testing, be it with a fix for the FTBFS on kFreeBSD on top of 1.0.42 or with a (unlikely) revert to 1.0.41, introducing an epoch. > ** Strictly speaking this is a rc bug in testing though one that is > probablly not worth filing unless one thinks the migration of the > missing build-dependency is unlikely in the forseeable future) I wasn't aware of that bug. Also, Debian hasn't usually made sure that testing was self-compilable outside of freezes, so as long as this gets fixed through unstable, I don't care too much (yet). Cheers, OdyX -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/11859141.KMjl9Jeq9V@gyllingar
