ModelingToolkit Integration

[xtalax]

Hey,
We're currently discussing requirements for the ModelingToolkit PDESystem frontend, and are building out tools to make it easier to integrate your package with this frontend. You already return SciMLProblems, so you would fit right in to the discretize interface.

see here for some discussion on the frontend

Please do reply if you are interested in hooking this package in to this

[DanielVandH]

Hi @xtalax: I was just reading that thread yesterday! I'm definitely interested in what you're doing over there. ModelingToolkit is especially something I initially wanted for this package, so having it set up would be amazing. Being able to have a better representation for more general domains for PDEs in general would also be great. Let me know if you want any other input.

[xtalax]

Really I'd just like to know, what exactly do you need to know about a system in order to discretize it? What sort of domains can you deal with?

Biggest question - what form do you need to be able to replace various derivatives with a finite volume discretization?
What do you need to do to your intermediate representation at the end?

Please weigh in on PDEBase, and let us know what you need.

[DanielVandH]

I see. I will give a reply to that thread you linked later today @xtalax, thanks. I think you forgot to link something at the end of this comment too?

[xtalax]

I got sidetracked, this is the issue you are looking for, especially the examples on tools given in the 3rd paragraph SciML/PDEBase.jl#8