Hi, we have an in-tree variant of the PicoSAT because the newer, reentrant version of picosat is around 30% slower for our usage than the one we're currently using (936). However, the picomus export is done for compatibility reasons. The output format was changed several times and the version 954 and 959 have some problems with argument passing and reading from stdin. These problems are solved with 960 but 960 was just released a few weeks ago. So we could use either 936 or 960 and i thought hat was a elegant way to solve the problem. But i didn't think it might cause conflicts here.
Stefan 2014-11-08 16:18 GMT+01:00 Reinhard Tartler <siret...@tauware.de>: > On 2014-10-21 19:53, Michael Tautschnig wrote: >> >> Hi undertaker maintainers/uploaders, >> >> [...] >>> >>> Here is a list of files that are known to be shared by both packages >>> (according to the Contents file for sid/amd64, which may be >>> slightly out of sync): >>> >>> /usr/bin/picomus >>> >> [...] >> >> I wasn't even aware that undertaker was in Debian, and the same might hold >> for >> you as far as picosat is concerned. Anyway, in the interest of code >> de-duplication, could undertaker be changed to use the separate picosat >> package >> instead of the one shipped as part of the source tree? > > > That was the case when we initially uploaded undertaker to debian AFAIR (or > it was the plan at least). We had to abandon this scheme because we switched > to a newer version of picosat that changed API. > > Stefan you just finished your Master thesis that worked a lot on how > undertaker uses the sat checker. Maybe you can give us some insight here? > > Thanks, > Reinhard -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org