[Dune] do intersections need to match?

[Martin Nolte]

Hi Christian,

If I remember correctly, we decided in Aachen (2013) that entities should have a 
unique orientation, i.e., the entity enumerates its corners in the same order 
independently from the way it was constructed.

Unfortunatly, the grid check still enforces the contrary: An entity returned by 
the method subentity must enumerate its corners in the same order prescribed by 
the concatenation of the element geometry and the reference embedding. In the 
case of ALUGrid, this enforces faces to have a different orientation when 
obtained from different sides of a conforming intersection.

As far as I remember, the question whether conforming intersections should have 
the same geometry as the face they coinside with (assuming that the face has a 
unique orientation) has not been considered so far. Personally, I would say it 
has to (otherwise I would consider it non-conforming).

Best,

Martin

On 03/05/2015 08:12 PM, Christian Engwer wrote:
> On Thu, Mar 05, 2015 at 06:14:00PM +0100, Aleksejs Fomins wrote:
>> Dear Dune,
>>
>> When a geometry of a face is requested as an intersection from two neighbouring elements, does the orientation of this face have to match? If yes, how is this orientation usually established in present grids.
>
> face don't have to match, but intersections are unique. If the mesh is
> conforming, the orientation of the faces, which co-incide with an
> intersection, can have a different orientation than the other face and
> than the intersection.
>
>>
>> So far I was naively using ReferenceElement.subEntity(...) to find the indices of a face from the containing element, but, naturally, there is nothing to enforce that the faces generated this way would have the same orientation...
>>
>> Regards,
>> Aleksejs
>>
>> _______________________________________________
>> Dune mailing list
>> Dune at dune-project.org
>> http://lists.dune-project.org/mailman/listinfo/dune
>>
>

-- 
Dr. Martin Nolte <nolte at mathematik.uni-freiburg.de>

Universität Freiburg                                   phone: +49-761-203-5630
Abteilung für angewandte Mathematik                    fax:   +49-761-203-5632
Hermann-Herder-Straße 10
79104 Freiburg, Germany