In the next release (or beta release) this will be fixed. So you can use common_refinement to intersect arbitrary fans (even with lineality). If you need this right away, use the attached patch and recompile polymake. Meaning: save the attached file somewhere. Change into the polymake trunk director...

Thank you! So I will ignore the part in the documentation that says: LINEALITY_DIM: common::Int Dimension of LINEALITY_SPACE. FIXME currently polymake can only handle POINTED cones Can you post the link to that part of the documentation? I hope that polymake will give me an error or warning if I us...

This bug has been fixed in the newest perpetual beta version. click

EDIT: if you downloaded the beta yesterday, please update your working copy with an svn up, since there where some difficulties.

This bug has been fixed in the newest perpetual beta version. click

EDIT: if you downloaded the beta yesterday, please update your working copy with an svn up, since there where some difficulties.

Today a new beta version is available (20131128). Just update your svn working copy (don't forget to configure and compile). More informations can be found here: http://www.polymake.org/doku.php/perpetualbeta EDIT: if you downloaded the beta prior to 29.11.2013 around 10:10am, please update your wor...

Hi there, no need apologizing for asking a question :-) The output is indeed not right. One should expect the following: fan > $p = new PolyhedralFan(INPUT_RAYS => [[1,0],[1,3]], INPUT_CONES=>[[0,1]]); fan > print $p->F_VECTOR; 2 1 fan > print $p->CONES; <{0} {1} > <{0,1} > Since the apex is the onl...

They are supported. You can define non-pointed fans... meaning that each cone of the fan has the same LINEALITY_SPACE. But there are some algorithms which have the precondition that the input should be a pointed fan. Often those algorithms can be generalized for the non pointed situation, but it jus...

Hi Hendrik, this was indeed a bug in the totally_dual_integral client. So thank you for finding it. I fixed that and it will be available in the next beta, which will be out soon. The problem had something to do with unbounded polyhedra. So make sure that you have 4ti2 configured to handle unbounded...

As far as I know, no one in the polymake team is using 10.9. This will probably change, but I don't know when they decide to update. Maybe one of the current Mac users can still troubleshoot your problem.

If you want to use ''is_totally_unimodular'', then yes (at least for now). But as I said you can use the function ''totally_unimodular'' The comment of this function is: # The matrix //A// is totally unimodular if the determinant of each square submatrix equals 0, 1, or -1. # This is the naive test ...