Benjamin Rodenberg

> The refactoring https://github.com/precice/precice/issues/1453 sadly turned out to be incorrect for implicit schemes https://github.com/precice/precice/issues/2022. Is there a possibility to test this to avoid this pitfall in the future? Or does...

Can you give a concrete example here? I have a few ideas what you mean and I think there is indeed some inconsistency, but I think it's easier to discuss,...

Potential preparation for this step: Change https://github.com/precice/precice/blob/e5a8cdaec1f700c1a95627d2ababa32188fb7012/src/cplscheme/BaseCouplingScheme.cpp#L120-L150 to ``` const auto &stamples = data->stamples(); PRECICE_ASSERT(!stamples.empty()); Eigen::VectorXd times; if (data->exchangeSubsteps()) { times = data->timeStepsStorage().getTimes(); } else { times timeStepsStorage().getTimes().getEnd(); // single-element...

Thanks for investigating. The source code you are pointing to above is clearly buggy. Let's fix this first in an independent commit. Can you create a PR that only contains...

I investigated the bug now a while to understand it better and came to the following conclusion: The following error ```diff - v_new = update_v(u, u_old, v_old, a_old) + v_new...

@fsimonis @IshaanDesai is this issue still relevant? I think looking at our other examples and solverdummies and how we make them available to the users (clone from github) we do...

From the following two perspectives I support this idea: * pySDC is very useful for **short-lived projects such as Bachelor or Master thesis projects** if they want to use advanced...

Possible solution: Some kind of mocking, like in [`fenics-adapter`](https://github.com/precice/fenics-adapter#test-the-adapter), where `SolverInterface.hpp` is mocked. It is sufficient to mock each API function separately, because we do not want to test preCICE...

C++ mocking framework: https://github.com/eranpeer/FakeIt

#420 and #431 implement automated testing for the python bindings.