Oliver Stöneberg
Oliver Stöneberg
I did not blindly add the parameters as it would change the behavior. This needs to be evaluated for each case if this was intentional or not.
> I think we should always pass `settings` if available, since that should only improve the results. Alright. Will do. I was thinking about setting breakpoints to the lines where...
It turns out this has quite a ripple effect so we should probably do that in a separate PR.
> It turns out this has quite a ripple effect so we should probably do that in a separate PR. See #6298.
I will re-structure these PRs.
There is most likely some explicit flushes to be added to have a good CLI experience. I have not looked into it yet.
It seems the missing flushes somehow expose a bug in the pipe communication in the `ProcessExecutor`: ``` reportOut - Checking /tmp/pytest-of-user/pytest-8/test_markup_j0/test_2.cpp ... handleRead - Checking /tmp/pytest-of-user/pytest-8/test_markup_j0/test_2.cpp ... reportOut - Checking...
I cannot reproduce it outside of the Python test so it must be related to `subprocess.Popen()`.
This totally throws off several premium-only checks...