[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

libode ABI breackage



Hi,

libODE (libode0-dev, libode0c2) is a free, industrial quality library for simulating articulated rigid body dynamics - for example ground vehicles, legged creatures, and moving objects in VR environments. It is fast, flexible, robust and platform independent, with advanced joints, contact with friction, and built-in collision detection. It is used by many games, as the trends are to provide more realistic behaviour of the engines every time. While version 0.5 has been in the archive for a while, most of the newer games demand version 0.8.

There is an ABI breakage between de 0.5 and the 0.8 release of libode. In fact, there are numerous reports of incompatibilities between versions in the forums, and the upstream developers apparently don't care at all about SONAMEs ( [WWW] http://bugs.debian.org/379791 ). Upstream are using the version number as a SONAME, so it seems they have no clue on what they're doing. Even though 0.5 and 0.8 are clearly not ABI-compatible (segfault), the major number of the SONAME seems to be the same.

Trying to upgrade to 0.8, and not being fully aware of the problem. we uploaded a newer version of libode0c2, based on the 0.8 version of ode. It seems to have been a bad idea, since programs depending on it seem to break. Thus, we wanted to ask what would be the best way to face this transition. We should also contact the maintainers of the packages that depend on this to try to make the transition as easy as possible for everyone.

Packages that depend on libode0c2: python-pyode, xmoto,  python-soya,  python-pyepl, and indirectly:

$apt-cache --recurse rdepends libode0c2 libode0c2 Reverse Depends:

python-pyepl Reverse Depends: python-soya Reverse Depends:

balazar Reverse Depends: python-soya-doc Reverse Depends:

slune Reverse Depends:

junior-games-gl Reverse Depends:


How is the proper way to handle this?

Greetings and thanks,
Miry

Please CC me, I'm not subscribed to the list


Reply to: