On Tue, 23, Aug, 2005 at 05:05:26PM -0700, Steve Langasek spoke thus.. > Please explain why you think this is a bug at all to allow this, let alone a > bug that warrants a release-critical severity. Nothing in your bug report > indicates that anything has actually *broken* by allowing these packages to > be co-installed.
Apologies. I appear to have made a mistake. This was brought to my attention by linda and lintian whilst getting kst ready for an upload for the G++ transition (I was ending up in some circumstances with ldd showing libstdc++5 and libstdc++6 dependencies despite the pbuilder sid built package only depending on libstdc++6). I think what happened is that FAM has C++ symbols but only exports a C API: http://lists.debian.org/debian-release/2005/08/msg00007.html so although it shows as broken (I was under the impression that linking against two versions of libstdc++ was normally a large bug), it's just because some of the private symbols in FAM are C++. Sorry for the noise, I think this can be closed now. Mark -- Mark Hymers <[EMAIL PROTECTED]> "The first thing we do, let's kill all the Lawyers" Henry VI Part II, Shakespeare
signature.asc
Description: Digital signature
