> apt-listbugs alerted me to this bug when I attempted to upgrade from
> 1.2.0-1.4. I can't tell from the information provided: if I upgrade to
> an affected version, should I expect things on my system to break?

I pinned the package, subscribed the bug and think it's good to have
some traffic in here. (Which of course will not fix the problem)
With a bit of luck, someone with knowledge tells ushow a proper solution
would look like.

Maybe this is useful, anyway there is a lot to read and learn:
* http://kernel-handbook.alioth.debian.org/ch-versions.html
* http://ispras.linuxbase.org/index.php/ABI_compliance_checker
* http://wiki.debian.org/Teams/ReleaseTeam/Transitions
* http://wiki.debian.org/Teams/ReleaseTeam

> Abi compliance checker
> http://ispras.linux-foundation.org/index.php/ABI_compliance_checker
> abi-compliance-checker can be used to help reduce the possibility of
> an application crashing when the shared libraries it links against
> have changed. Signatures and data type definitions from two separate
> versions of a library are compared by examining the shared objects
> (.so) files themselves, and by analysing the header files (.h)
> provided for the two versions of the library and their dependencies.
> Shared library developers trying to strive for binary compatibility
> between releases may also use this tool to detect if an any
> accidental application binary interface (ABI) changes have been
> introduced. The checker may also be used for assessing binary
> compatibility between different Linux distributions.


gpgme1.0 (1.4.1-0.1) unstable; urgency=low

  * debian/control: Enable multiarch support (closes: #698970).
    (Priority): Raised to standard (closes: #623353).
    (Section): Changed to libs. Fixed
    binary-control-field-duplicates-source. (Standards-Version): Bumped
    to 3.9.4. (Vcs-Git, Vcs-Browser): Adjusted to point to svn tree
    (closes: #610737). (Build-Depends, Depends): Added new dependencies
    including libassuan-dev. Fixed debhelper-but-no-misc-depends.
    Dropped dpatch and obsolete libpth-dev. Replaced gnupg by gnupg2.
    (Description): Fixed several lintian hints.
  * debian/libgpgme11.files: Renamed to debian/libgpgme11.install,
    adjusted.
  * debian/libgpgme11-dev.files: Likewise. Dropped .la (LP: #728497).
  * debian/libgpgme11.dirs, debian/libgpgme11-dev.dirs: Dropped useless
    files.
  * debian/libgpgme11-dev.info: Ditto.
  * debian/libgpgme11-dev.doc-base (Section): Fixed.
  * debian/libgpgme11.symbols: Updated. Removed symbols of
    libgpgme-pth.so.11.

 -- Daniel Leidert <dleid...@debian.org>  Tue, 14 May 2013 20:29:20
 +0200


< http://wiki.debian.org/ReleaseTeamOverloaded
I think that part of the problem is that we're leaning on the release
team to make hard decisions that we should be capable of making
ourselves. -- JoeyHess

Read debian-release, you'll frequently see developers:
* Say: "can this <random change> get into sarge?"
* Trying to get in a change without thinking it through.
* It's also annoyingly common for maintainers to upload a fix that
really needs to get into sarge with other nonessential changes that
increase the probability a bug will be introduced
* Upload something such as a security fix with a priority that's too low
* Package a new upstream version of a library in a way that it holds
lots of other things out of sarge.

All of these are responsible for a lot of the load on the release team.
All can be avoided by maintainers thinking through the conseqences. --
JoeyHess


-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Reply via email to