Sam Hartman <[email protected]> writes: >>>>>> "Martin" == Martin Pitt <[email protected]> writes:
> Martin> So right now, it seems that reintroducing a transitional > Martin> libkrb5 metapackage to pull in the others is a rather safe > Martin> and robust way, and it should also help apt & friends. What > Martin> do you think? > I'm not sure. We considered and rejected that option but I'm not sure > why rejected it. It may simply have been that we didn't see value in > the transitional package and that technically speaking the transitional > package should not exist because it dosen't actually provide the > libkrb53 ABI. As I recall, that's why we rejected it: we can't build a transitional package that provides the same ABI as libkrb53 in lenny, so traditionally one is not supposed to use a transitional package in this case. Note that taking this approach will break any out-of-archive packages that rely on the libkrb53 ABI. > My conclusions so far are: > 1) I don't see why the transitional package is not the best solution we > have on the table I agree. -- Russ Allbery ([email protected]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
