Conversation
b1e5eff to
3ea0bde
Compare
|
So i had a look at the problem at hand and could identify two things:
I will try to add reference results now via the pipeline. |
|
Another problem i found is that we use the v202211.0 tag of the tutorials as a reference. But there the CPP code is not built properly, but assumed to be built manually. (see https://github.com/precice/tutorials/blob/v202211.0/elastic-tube-1d/solid-cpp/run.sh). So its gonna be tricky to merge this somewhere soon, because we cannot really generate reference results for it? One way forward would be to cherry pick the commit into the master, before the release? |
Good catch! Since the new |
3ea0bde to
e30df9b
Compare
|
dc7335a solved the problems. The VM was able to add the reference results, so i guess we can merge this now. |
MakisH
left a comment
There was a problem hiding this comment.
The generated results contributed here look reasonable (I see the pressure and cross-section length oscillations).
Adds a test suite for
elastic-tube-1d, which is a special case, since it does not rely on any adapter (but relies partially on the python-bindings).Overview of changes:
bare, which essentially only relies on thepreciceDocker image.python-bindingsmetadata.ymlinto theelastic-tube-1dtutorialelastic_tube_1d_test, with the tests:However, only the last test is being executed: #394
Further observations
This test correctly fails to build the code and run the simulation when executing
generate_reference_data.py, since the tutorial is not yet ported to preCICE v3 (#395). However, running the system tests only complains at the stage of comparing the results: