[BeamInterpolation] Generalize parameters per edge #169
Conversation
fredroy
added
the
pr: status to review
fredroy
changed the title
fredroy
changed the title
|
I found a problem. I'll remove the draft status when I'm done. EDIT - IMPORTANT: My problem was due to the usage of |
|
Hey @EulalieCoevoet |
|
Yes sorry. I just fixed it and added more tests. |
| Data<helper::OptionsGroup> crossSectionShape; | ||
|
|
||
| /// Circular Cross Section | ||
| Data<Real> d_radius; | ||
| Data<Real> d_innerRadius; | ||
| Real m_defaultRadius; |
There was a problem hiding this comment.
It seems to be lost in the report but it has been noted that it seems all the m_default* are constant (and even static) so they should be mark as constexpr (and initialized inline)
| checkDataSize(m_defaultRadius, d_radius, nbEdges); | ||
| checkDataSize(m_defaultInnerRadius, d_innerRadius, nbEdges); | ||
| checkDataSize(m_defaultLengthY, d_lengthY, nbEdges); | ||
| checkDataSize(m_defaultLengthZ, d_lengthZ, nbEdges); | ||
| checkDataSize(m_defaultYoungModulus, d_defaultYoungModulus, nbEdges); | ||
| checkDataSize(m_defaultPoissonRatio, d_poissonRatio, nbEdges); |
There was a problem hiding this comment.
These calls change the m_defaultvalues based on the d_values set by the user.
See line 193 :
defaultValue = values[0]; // if the sizes mismatch again at runtime, will use this default value
We can already set a
defaultYoungModulusanddefaultPoissonRatioper edge.This PR generalizes this option to all parameters:
radiusinnerRadiuslengthYlengthZmassDensity(AdaptiveBeamForceFieldAndMass.inl)Question: Should we remove the options
sideLength,largeRadiusandsmallRadius? They are not used, since square and elliptic cross-section are not implemented.EDIT: If you agree, we should also remove the elliptic and square options from
Data<OptionsGroup> crossSectionShape.