El Dijous, 6 d'octubre de 2016, a les 15:23:10, Ponomarenko Andrey va escriure: > 06.10.2016, 07:54, "Paul Wise" <email@example.com>: > > On Wed, Oct 5, 2016 at 11:00 PM, Ponomarenko Andrey wrote: > >> I'd like to present a new free tool for maintainers of software > >> libraries — Package ABI Diff Tool (Pkg-ABIdiff). It's a tool for > >> backward compatibility analysis of API/ABI interfaces in Deb packages. > >> The tool is based on ABI Compliance Checker and ABI Dumper tools.> > > Does this have any advantages over abipkgdiff from the abigail-tools > > Debian package already in Debian? > > > > BTW, I think it would be really interesting to run > > pkg-abidiff/abipkgdiff over the whole Debian archive and possibly use > > that to inform the release team of uncaught ABI changes. > > > > -- > > bye, > > pabs > > > > https://wiki.debian.org/PaulWise > > I guess, checks for more compatibility rules, less false positives, visual > reports, problem severity levels and separated analysis of both backward > binary compatibility and backward source compatibility are main advantages > of the pkg-abidiff. The disadvantage is that pkg-abidiff may not be as fast > as abipkgdiff, because pkg-abidiff is written in Python, but abipkgdiff is > written in C++. > > The tools are based on different software stacks. The pkg-abidiff is based > on ABI Compliance Checker and ABI Dumper tools (https://github.com/lvc) > developed since 2009. The abipkgdiff is based on libabigail developed since > 2013. So, implementations and reports are completely different. > > Anyway, it's better to run both tools on all packages at the same time and > verify reports of each other. > Then I would like to ask when we must think that we need a transition for a the package. If these test shows a binary compatibility of 99%, do we need to create a need soname bump and initiate a transition? Best regards, Leopold -- -- Linux User 152692 GPG: 05F4A7A949A2D9AA Catalonia ------------------------------------- A: Because it messes up the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing in e-mail?
Description: This is a digitally signed message part.