WIP: Add compat for new AbstractCell type. #78
Conversation
|
I think the new interface will be a win in FerriteViz too, most dispatches seem to be based on the reference shape, which you can now to directly instead of figuring out the correct numbers in |
|
I introduced the abstract methods for now to have at least some kind of viz. Actually the high order stuff needs quite a different treatment but I could not find time yet. |
| @@ -10,65 +10,65 @@ for_nodes(::Union{Ferrite.Lagrange{2,Ferrite.RefTetrahedron,2},Ferrite.Discontin | |||
| (3,6,5), | |||
| (5,4,2), | |||
| ) | |||
| # Cubic Triangle | |||
| for_nodes(::Union{Ferrite.Lagrange{2,Ferrite.RefTetrahedron,3},Ferrite.DiscontinuousLagrange{2,Ferrite.RefTetrahedron,3},Ferrite.Cell{2,10,3}}) = ( | |||
There was a problem hiding this comment.
Note that this also includes the refinement for interpolations. Maybe we just want to remove the Cell from the type union for now. Same for ones below.
| ntriangles(cell::Ferrite.AbstractCell{3,N,6}) where N = 6*4 # Hex | ||
| ntriangles(cell::TriangleBasedCells) = 1 # Tris in 2D and 3D | ||
| ntriangles(cell::QuadrilateralBasedCells) = 4 # Quads in 2D and 3D | ||
| # ntriangles(cell::Ferrite.AbstractCell{3,N,1}) where N = 4 # Tets as a special case of a Quad, obviously :) |
There was a problem hiding this comment.
Can be removed now that embedded cells do not have the case distinction anymore.
| # ntriangles(cell::Ferrite.AbstractCell{3,N,1}) where N = 4 # Tets as a special case of a Quad, obviously :) |
|
I am donating this PR to someone who knows better :) The current changes are mostly mechanical without knowing much how these things are used. |
|
Will take over then. |
|
Handled in #103 |
No description provided.