check_isoparametric_boundaries, mass_qr, and cellcenter for Pyramids and Wedges + L2Projector tests for Hexahedron, Tetrahedron, Pyramid, and Wedge.#1215
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #1215 +/- ##
==========================================
- Coverage 94.25% 93.07% -1.18%
==========================================
Files 40 40
Lines 6750 6859 +109
==========================================
+ Hits 6362 6384 +22
- Misses 388 475 +87 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
| [deps] | ||
| EnumX = "4e289a0a-7415-4d19-859d-a7e5c4648b56" | ||
| ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210" | ||
| Interpolations = "a98d9a8b-a2ab-59e6-89dd-64a1c18fca59" |
lijas
left a comment
There was a problem hiding this comment.
Is it easy to implement the linear interpolation ourself and avoid importing the Interpolations package?
| function test_projection(order, elementtype) | ||
| refshape = getrefshape(elementtype) | ||
| dim = Ferrite.getrefdim(refshape) | ||
| grid = single_element_grid(elementtype) |
There was a problem hiding this comment.
Why did you change this to a single element grid? Previously we had triangles/tetrahedrons filled a square/cube.
There was a problem hiding this comment.
My bad, I didn't notice it.
| test_projection(2, RefQuadrilateral) | ||
| test_projection(2, RefTriangle) | ||
| for ref_shape in ( | ||
| Quadrilateral, |
| # A linear approximation can not recover a quadratic solution, | ||
| # so projected values will be different from the analytical ones | ||
| ae = [f_approx(i) for i in 1:4] | ||
| qp_1D_coord = Float64[] |
There was a problem hiding this comment.
Add a small comment what you are trying to do here
There was a problem hiding this comment.
I was trying to calculate the first order approximation values using linear extrapolation from the quadrature points instead of using precomputed hardcoded values. But since we use more than one element this won't work so I'll compute the values for other cell types and add them instead.
…mids and prisms need to be increased so that the matrix is matrixing
check_isoparametric_boundaries and cellcenter for Pyramids and Wedges + better tests for L2Projector.check_isoparametric_boundaries, mass_qr, and cellcenter for Pyramids and Wedges + L2Projector tests for Hexahedron, Tetrahedron, Pyramid, and Wedge.
The new functions are never called in the tests. |

PointEvalHandler needed these functions to work. Also, L2 Projector is only tested with triangles and quads. I added a bit more general test using
Interpolations.jlinstead of manually calculating the test values for first order interpolations.