[Dune] Gmsh reader and boundary Ids

[Oliver Sander]

> As of now I have not noticed much pattern in the boundary segment index provided by individual grids.

For UGGrid the following should hold: if you (i.e., the GmshReader) explicitly inserts boundary segments,
then the value returned by boundaryIndex reflects the insertion order.  If you do not explicitly insert
the boundary segments, then the ordering of the values of boundaryIndex is practically arbitrary.
--
Oliver
> 
> P.S. In the curvilinear grid we are currently developing, it will be possible to extract the PhysicalID from the intersection, as it will be stored inside the grid after reading GMSH.
> 
> Cheers,
> Aleksejs
> 
> 
> 
> 
> On 14/04/15 17:17, Agnese, Marco wrote:
>> Hello Duners,
>> I am reading a mesh doing
> 
>> Dune::GmshReader<GridType>::read(gridFactory,fileName,boundaryPhysicalIDs,elementsPhysicalIDs);
> 
>> If I want to extract the ID of an intersection I need to do
> 
>> boundaryPhysicalIDs[gridFactory.insertionIndex(intersection)]
> 
>> while, looking at the documentation, it seems natural to do
> 
>> boundaryPhysicalIDs[intersection.boundarySegmentIndex()];
> 
>> It wouldn't be better to reorder inside the gmsh reader the boundaryPhysicalIDs in order to be consistent with the boundary segment index?
>> I know that the reorder takes only 5 lines of code but it would be more user friendly to have it done internally.
> 
>> Maybe I am missing the purpose to have a vector ordered following the insertion index.
> 
>> Cheers,
>> Marco.
>> _______________________________________________
>> Dune mailing list
>> Dune at dune-project.org
>> http://lists.dune-project.org/mailman/listinfo/dune
> 
> 
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune
> 


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <https://lists.dune-project.org/pipermail/dune/attachments/20150414/e9ec6e56/attachment.sig>