Allow ForceModels to accept any number of models
#262
Conversation
|
Forgot we'll need to bump the Cabana hash in the CI/README |
|
@streeve can we run the CI on this? I think it is at least in a state where it can work |
streeve
left a comment
streeve
left a comment
There was a problem hiding this comment.
This is great; I think the only real question is whether we want to add more testing before merge. @pabloseleson do you have a clear idea of a binary material test system that we can easily check forces/energies for in a similar way we currently do single materials (could be a separate PR)?
| KOKKOS_ASSERT( firstType < NumBaseModels ); | ||
| KOKKOS_ASSERT( secondType < NumBaseModels ); | ||
|
|
||
| return firstType != secondType; |
There was a problem hiding this comment.
how about a ternary? I think it is easier to read:
return (firstType != secondType) ? 1 : 0;
JBludau
force-pushed
the
revisit_ForceModelMulti
branch
from
457b0e0 to
4e891e9
Compare
|
again? |
|
git wanting to go on holiday |
JBludau
force-pushed
the
revisit_ForceModelMulti
branch
from
e577ed9 to
9d847b1
Compare
JBludau
force-pushed
the
revisit_ForceModelMulti
branch
from
9d847b1 to
858b01b
Compare
| static_assert( | ||
| (std::conjunction_v< | ||
| std::is_same<typename ParameterPackType::template value_type< | ||
| Indices>::model_tag, |
There was a problem hiding this comment.
We could instead check force_tag which is slightly more restrictive. Or we can check both to be more certain if things change in the future
There was a problem hiding this comment.
I am for being more restrictive ... it is easier to loosen the restrictions later than to tighten them. And these asserts are mainly for our own protection
2 participants
This is a first template for how a the Multiforce model can be changed so it accepts a unspecified number of models.
It requires some meta-template tinkering and some work still:
ParameterPackto be constructed from elements that are not default constructible ECP-copa/Cabana#855