On Oct 19, 4:37 am, Volker Braun <vbraun.n...@gmail.com> wrote: > Yes, PALP has some useful toric functionality and I think we should > definitely keep it around. But I think the long-term goal should be to > reduce the reliance of the LatticePolytope class on PALP. In > particular enumerating the lattice points is a problem that pops up > frequently enough that an implementation without the dimension > limitations of PALP would be nice. Did you ever figure out which > algorithm PALP uses? I also looked at the source code but I quickly > gave up...
No, I didn't. Although as far as I know I read all available documentation on it. I think that so far I needed lattice points only inside reflexive polytopes in dimension up to 6 and in this case PALP limitations do not show up. I also didn't have any need to do it for sequences of polytopes where next elements are computed based on previous ones, so my work-around (all_points) to eliminate extra system call overheads was sufficient. By the way - does PPL use any random numbers like cddlib? I like referencing polytope vertices/points/facets by their numbers and it will be impossible if the order can change from session to session. I don't care about having some specific order - just a fixed one (except that vertices should be listed in the same way in which they were given by the user, I think). -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
