On 02/03/2015 11:59 AM, David Bremner wrote: > Tobias Hansen <[email protected]> writes: >> >> Hi David, >> >> The problem is that Sage patched libcdd and added features that are not >> included upstream [1]. The changes are actually not that big. The new >> features are in cdd_both_reps.c and the rest should be mostly a switch >> of the random number source to something deterministic, I guess to avoid >> problems with tests. > > I haven't had a chance to look at the changes in cdd_both_reps.c, but I > wonder if they really need to be in cddlib, or if this just seemed > easiest at the time. Anyway, it's on my list of things to think about; > polymake, which I'm in the process of packaging, also embeds a copy of > cddlib; in that case it needs to link to the library compiled two > different ways: floating point and gmp arithmetic. > > Alas, no definite schedule from my end. > d
Both versions of cddlib are in the package (libcdd0d) so that should work for polymake, right? You are right, maybe cdd_both_reps.c could become a new library. Best, Tobias -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: https://lists.debian.org/[email protected]
