converting std::vector<double> to std::vector<float> for BIC coefficients in GenericOilGasWaterFluidSystem

[GitPaean]

for testing the PR OPM/opm-simulators#7049 .

[GitPaean]

jenkins build this please

[GitPaean]

It is used to compile the float version of flow variants.

https://ci.opm-project.org/job/opm-simulators-PR-builder/9985/console

But it is also possible the refactoring in OPM/opm-simulators#7049 should not have triggered this issue.

[Copilot]

Pull request overview

Updates GenericOilGasWaterFluidSystem initialization to allow BIC (binary interaction coefficient) data coming from CompositionalConfig (std::vector<double>) to be stored in the fluid system’s interaction_coefficients_ (std::vector<Scalar>), enabling builds where Scalar is float (as needed for the referenced simulator PR).

Changes:

• Replace direct assignment of BIC (std::vector<double>) into interaction_coefficients_ with an iterator-based copy to support differing element types (double → Scalar).
• Introduce a local bic reference for the retrieved BIC vector.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[GitPaean]

It is used to compile the float version of flow variants.

https://ci.opm-project.org/job/opm-simulators-PR-builder/9985/console

But it is also possible the refactoring in OPM/opm-simulators#7049 should not have triggered this issue.

Without looking into details yet, since OPM/opm-simulators#7049 is trying to make compositial based on the Flow framework, when the instantiation based on the GenericOilGasWaterFluidSystem for FlowGenericProblem might be necessary.

[GitPaean]

jenkins build this please

[GitPaean]

It is needed if (-DBUILD_FLOW_FLOAT_VARIANTS is on) through OPM/opm-simulators#7049 .

[GitPaean]

should we do it through different way? @bska @akva2

[akva2]

Looks fine to me

[GitPaean]

jenkins build this please

[akva2]

though clang might not agree, so maybe something akin to https://github.com/OPM/opm-simulators/blob/master/opm/simulators/flow/equil/InitStateEquil_impl.hpp#L1528 is necessary to shut it up

[bska]

Can we at least keep the current assignment if Scalar is double? For instance, something along the lines of

if constexpr (std::is_same_v<Scalar, double>) {
    this->interaction_coefficients_ = ...;
}
else {
    const auto& bic = ...;
    this->interaction_coefficients_.assign(bic.begin(), bic.end());
}

with std::is_same_v<> from the <type_traits> header.

[GitPaean]

I combined both comments, please let me know if it is okay.

Implicit conversion from double to float will always be there with either approach?

[GitPaean]

jenkins build this please

[bska]

Thanks a lot for the updates. This looks good to me now and I'll merge into master.