Subject: CGAL users discussion list
List archive
- From: "Brown, James 2" <>
- To: "''" <>
- Subject: [cgal-discuss] Changed cells in Delaunay_triangulation_3
- Date: Wed, 31 Aug 2016 15:50:41 +0000
- Accept-language: en-US
- Authentication-results: mail2-smtp-roc.national.inria.fr; spf=None ; spf=Pass ; spf=Pass
- Ironport-phdr: 9a23:Kt/CmhCqx+Tqer3Zt0sjUyQJP3N1i/DPJgcQr6AfoPdwSP77oMbcNUDSrc9gkEXOFd2CrakV0qyK4+uwAiRAuc/H6y9SNsQUFlcssoY/oU8JOIa9E0r1LfrnPWQRPf9pcxtbxUy9KlVfA83kZlff8TWY5D8WHQjjZ0IufrymUt2as8Pina+59JTXJglJnzGgeqhaLROsrAyXuNNZy8M2Iak4zl7Fo2BDZv9N7WJuP1Oa2RjmsJSe5plmpm56ofU+8YoIc6z/cr9wF/YQWDYvNm0uosfxsgbFZQ+K7XxaWWITxEkbSzPZ5Q33C8+i+hDxsfBwjXGX
I need to find changed cells (removed and new) for moved vertices. The movement I have is fairly small but because some tetrahedra are very ‘thin’ sometimes the points cross a face causing the tetrahedron to be in an opposite orientation which would require the neighboring cells to be retriangulated.
Currently for insertions I am using find_conflicts and filtering on finite cells to find removed cells along with insert_in_hole for the insertion, and then using finite_incident_cells of the inserted vertex as the new cells. This seems to do the job but please correct me if this is wrong or there is a better way.
For moving a point it looks like if I could get access to the hole in move_if_no_collision it would give the removed cells, and then would getting the finite_incident_cells of the inserted (moved) vertex be sufficient in this case?
As well I noticed the x_and_give_new_cells series of functions that are marked internal, are these of any use?
Thanks, James |
- [cgal-discuss] Changed cells in Delaunay_triangulation_3, Brown, James 2, 08/31/2016
Archive powered by MHonArc 2.6.18.