Under some circumstances, when the ray from experimental data is aligned with the mesh, RayTracing::get_intersection can erroneously give DOF index 0 as the intersection point.
There's a note in the function that rays that cross an edge may be lost -- I think this is separate from that, although perhaps related. In the cases I saw, the ray was going through a part of the face away from the edge (and not the exact middle of the face either).
Under some circumstances, when the ray from experimental data is aligned with the mesh,
RayTracing::get_intersection
can erroneously give DOF index 0 as the intersection point.There's a note in the function that rays that cross an edge may be lost -- I think this is separate from that, although perhaps related. In the cases I saw, the ray was going through a part of the face away from the edge (and not the exact middle of the face either).
When I have a chance, I'll post a reproducer.