On Sat, 2009-01-17 at 16:40 +0100, Jonas Smedegaard wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On Sat, Jan 17, 2009 at 10:02:27AM +0100, Luca Falavigna wrote: > >James Westby ha scritto: > >> At the end of last year there was a small transition done to move > >> from libc-client2007b to libc-client2007d. > >> > >> However, Debian then reverted this change (with an epoch), as they > >> didn't want the SONAME bump in lenny. > > > >I'm not sure Debian reverted the upload only to avoid SONAME bump, > >there was probably a problem in the upstream code itself, but I'm not > >sure about this. Maybe Jonas could shed some light. > > There was no problems with upstream code, it was only to avoid SONAME > bump. > > Upstream has no shared library, that is a Debian hack. But historically > upstream license required (by my interpretation at least) the upstream > version to be clearly visible in derived code, which lead me to the > current unusual packaging routine of bumping SONAME with each new > upstream release. > > In other words: There is no technical reason for SONAME to be changed so > frequently. It was wrong of me to bump the SONAME of a library in deep > freeze, but when the "cat was out of the bag" it was suggested to change > epoch as a simplest workaround.
Hi Jonas, Thanks for the information. Do you think that Luca's suggestion to follow you in Debian and pull your revert of the SONAME, and then "un-transition" the packages that we had rebuilt for the new SONAME is a good one? Once lenny is released then we can perform the transition at the same time as you. Thanks, James -- Ubuntu-motu mailing list [email protected] Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-motu
