Fix the jobs status script so it works with the new gbasf2 version

[eckerpatrick]

There were some changes in the recent gbasf2 update which caused failures in the job status script, as the CS is not anymore automatically initialized by the userCreds decorator.

[eckerpatrick]

this fixes #208

[codecov-commenter]

Codecov Report

Attention: 5 lines in your changes are missing coverage. Please review.

Comparison is base (5815d46) 60.51% compared to head (fa7f42c) 60.87%.
Report is 15 commits behind head on main.

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #209      +/-   ##
==========================================
+ Coverage   60.51%   60.87%   +0.36%     
==========================================
  Files          23       23              
  Lines        1593     1595       +2     
==========================================
+ Hits          964      971       +7     
+ Misses        629      624       -5     

Flag	Coverage Δ
unittests	60.87% <54.54%> (+0.36%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files	Coverage Δ
b2luigi/__init__.py	88.46% <100.00%> (ø)
b2luigi/batch/processes/gbasf2.py	49.78% <50.00%> (+0.52%)	⬆️

... and 6 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[meliache]

Thanks, that looks great 👍 ! @0ctagon had figured out the changed import paths but was missing initializeCS().

If you don't mind I'll also add commit 4418938 for better catching unexpected errors of gbasf2_job_status.py. In the b2luigi in the subprocess call of gbasf2_job_status.py we only checked for a job status of 3, which happens when result["OK"] is False. However, any other unexpected crash of that script resulted in a non-zero exit code which b2luigi ignored and just continued with a garbage string. Therefore, we should also raise error messages for other non-zero exit codes. Not sure if that would have resulted in a better/earlier error message in this particular case though.

What's finally missing is a Changelog entry.

[eckerpatrick]

@meliache sounds good to catch also other failures!
Will you also add the changelog in this context or should I add it?

[meliache]

@eckerpatrick Oh you're quick, I just added a Changelog entry in addition to 4418938. Can you just make a quick check that my changes didn't brick anything? They are minor and shouldn't brake anything, but I have no way to run gbasf2 to check anymore and who knows what I might have overlooked.

(These days I started sometimes merging minor hotfixes/improvements without proper code review using my admin authority and because it'd difficult to find reviewers/testers these day. But people get active here when something is broken so I'm happy to exploit that.)