Add homogeneous Hermite to the possible spline boundary conditions#1062
Add homogeneous Hermite to the possible spline boundary conditions#1062tpadioleau merged 12 commits intomainCExA-project/ddc:mainfrom ebourne_spline_homogeneous_hermiteCExA-project/ddc:ebourne_spline_homogeneous_hermiteCopy head branch name to clipboard
Conversation
42fb141 to
25cb3ef
Compare
Codecov Report❌ Patch coverage is
❌ Your patch status has failed because the patch coverage (93.93%) is below the target coverage (100.00%). You can increase the patch coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## main #1062 +/- ##
==========================================
- Coverage 94.72% 94.70% -0.02%
==========================================
Files 59 59
Lines 2863 2873 +10
Branches 890 898 +8
==========================================
+ Hits 2712 2721 +9
- Misses 93 94 +1
Partials 58 58 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
77b676f to
f794691
Compare
f794691 to
ecb80d2
Compare
There was a problem hiding this comment.
Looks ok to me.
We discussed the possibility to generalize it to a constant value but it doesn't look trivial to extend the builder in that way. The constructor does not take any input data related to the "boundary conditions".
Also looking at this feature made me wonder whether we actually need some parameters as templates ? Are the current BoundCond and linear solver needed as templates ? Same question in the tests, I am not convinced we need a new executable per evaluator.
|
The coverage failure is ok, buggy toolchain. |
The most common use case (by far) for Hermite boundary conditions is that of homogeneous Hermite boundary conditions (i.e. the case when the value of the provided derivatives at the boundary is 0). It can be costly to create a ND
ChunkSpanto hold the values of the derivatives at the boundary. Adding a homogeneous Hermite boundary condition avoids this cost as the 0 can be hardcoded.