glib2.0 version is wrong

How is that a bug or an issue? It doesn't claim to supersede anything, that's just a sync on a debian svn version which was not uploaded to debian yet

the revision seems to be different from what has been uploaded to debian though, but still that should create no issue in ubuntu

The version number means "snapshot of -2 which will be supersedable by -2". It was taken when -2 was in the works and no "ubuntu" was added in the version number to allow superseding with whatever next Debian upload we like (final -2 for instance). I could as well use -1+hardy1 instead next time if this is clearer to you, but I usually use "next Debian rev" + "~hardy1" as it is consistent when I pull a new upstream (most common case).

I find it confusing because 2.16.1-2 is listed in the changelog (a 2.16.1-2~... version should happen before that). Additionally, it would result in some potentially confusing security update versioning. Also, people have come to expected "ubuntu" to always be part of the version string if there are ubuntu-specific changes. I don't mean to imply the current version is incorrect for technical reasons -- I just think it should be consistent with other ubuntu changes. Debian to Ubuntu merges are supposed to include standard text about "Merged with Debian, remaining changes:" etc, and have an ubuntu* version.

Hmm I could have used the same changelog entry instead of creating a new one indeed.

(Concerning security update versioning, I didn't understand what it changes.)

About the "ubuntu" version number, well it's exactly the point: this is not Ubuntu specific at all; it's just a snapshot of work in progress or of a package pending NEW in Debian for instance; there is /no/ Ubuntu change to carry, just a snapshot of the WIP package from Debian to get it faster. And we wont have any merge either.

Looking back at the changelog, I used -2~hardy1 while -2 was already in Debian; what I meant to use was -3~hardy1, which might explain why this was confusing. Another option would be -2+hardy1 which might be clearer.

On Tue, Apr 01, 2008 at 09:07:54AM -0000, Lo??c Minier wrote:
> Looking back at the changelog, I used -2~hardy1 while -2 was already in
> Debian; what I meant to use was -3~hardy1, which might explain why this
> was confusing. Another option would be -2+hardy1 which might be
> clearer.

Right, yes. -3~hardy1 makes total sense -- -3 didn't exist yet. :)

Thanks!

This bug was fixed in the package glib2.0 - 2.16.2-1ubuntu1

---------------
glib2.0 (2.16.2-1ubuntu1) hardy; urgency=low

  * Sync on Debian
    - use correct versionning (lp: #209308)
  * debian/patches/06_dont_list_userdir.patch:
    - don't list the userdir as a mount, fix nautilus and autofs issues
      (lp: #191512)

 -- Sebastien Bacher <email address hidden> Thu, 03 Apr 2008 22:15:57 +0200