You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I would like to experiment with the very exciting uv-theory stuff in my code, and I can call that from Python, but to integrate with the tools I am developing, I need a C/C++ interface. Is it possible to develop such an interface easily with rust?
Technically, it is possible. Practically it would involve a lot of work. There are tools to generate C/C++ header files from the Rust library (cbindgen) but - similar to the Python interface - we would need to define wrapper functions with C compatible data types.
The amount of work needed depends on what functionalities of the library you need. I can create an example repository that showcases how such an interface could be built for the specific use case in question.
Yes, this is what I figured/feared. I hoped that the bright, shiny rust might have magical capabilities that have eluded C++ thus far.
What I need is the core derivatives of the residual Helmholtz energy with respect to temperature and density, up to high degrees, most especially to calculate speed of sound. I'm trying to find a model that can give good representations of speed of sound as well as densities and pressure, with accuracies on par with the multiparameter EOS implemented in REFPROP/CoolProp. So far the "old SAFT" models don't meet the brief, but I am hopeful about the newer approaches; SAFT-VR-Mie is the best candidate of the SAFT models so far that are implemented in Clapeyron.jl.
Activity
g-bauer commentedon Mar 15, 2023
Technically, it is possible. Practically it would involve a lot of work. There are tools to generate C/C++ header files from the Rust library (
cbindgen) but - similar to the Python interface - we would need to define wrapper functions with C compatible data types.The amount of work needed depends on what functionalities of the library you need. I can create an example repository that showcases how such an interface could be built for the specific use case in question.
ianhbell commentedon Mar 15, 2023
Yes, this is what I figured/feared. I hoped that the bright, shiny rust might have magical capabilities that have eluded C++ thus far.
What I need is the core derivatives of the residual Helmholtz energy with respect to temperature and density, up to high degrees, most especially to calculate speed of sound. I'm trying to find a model that can give good representations of speed of sound as well as densities and pressure, with accuracies on par with the multiparameter EOS implemented in REFPROP/CoolProp. So far the "old SAFT" models don't meet the brief, but I am hopeful about the newer approaches; SAFT-VR-Mie is the best candidate of the SAFT models so far that are implemented in Clapeyron.jl.