On Tue, Aug 14, 2007 at 08:20:25PM -0400, Joey Hess wrote: > Steve Langasek wrote: > > I'm in total agreement with this. I was staying out of this thread because > > I've been one of the proponents of using -0.x for NMUs of native packages in > > spite of the inconsistency with Policy, and I wasn't sure that this > > reasoning wasn't just a post-hoc rationalization on my part. Since you've > > come to the same conclusion, I suppose it isn't. :)
> Since you've thought about this, did you also consider the case of > bin-NMUs of native packages? Is that even done? If it is, and the > version gets a dash added to it as part of the bin-NMU mangling, such a > bin-NMU could trigger #437392. Native packages do get binNMUed; anna was recently binNMUed on several archs in response to the ldbl symbol skew, so there are now 1.27+b1 packages in the archive for alpha, powerpc, s390, and sparc. So no dash being added to the version number, as none is needed, and in any case the binNMU handling that was spec'ed out wouldn't recognize a package as a binNMU if there were anything more than '+b<n>' added to the sourceful version number. (Although, 1.27+b1 sorts after 1.27-0.1, so you can't NMU a native package with this version scheme that has previously been binNMUed...) -- Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. [EMAIL PROTECTED] http://www.debian.org/ -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
