Properly unpack 0D data when reading an IDS from a netCDF file

[maarten-ic]

0D IDS Data is expected in the native python data types (int, float, complex, string). Before this fix, 0D numerical data would be stored as numpy.int32, numpy.float64 or numpy.complex128 when reading from a netCDF file.

Fixes #89

[maarten-ic]

@timothy-nunn would you be able to test this fix as well?

[olivhoenen]

Is this a follow-up from #57?
I'm a bit confused because I can't reproduce the error with the code snippet from the issue #89 (and I see no difference of behavior between 2.1.0 tag and this bugfix branch)

[maarten-ic]

Is this a follow-up from #57?

Yes, it is

I'm a bit confused because I can't reproduce the error with the code snippet from the issue #89 (and I see no difference of behavior between 2.1.0 tag and this bugfix branch)

That's interesting.. I can reproduce the issue on SDCC with IMAS-Python 2.1.0 (module load IMAS-Python). Could you try that as well?

[timothy-nunn]

Hi @maarten-ic,

I was able to reproduce the error on the latest main (72d7ea3).

This PR fixes the problem and my code snippet now runs with no errors.

Thank you very much for your prompt response and solution to this issue, it is much appreciated!

[olivhoenen]

Is this a follow-up from #57?

Yes, it is

I'm a bit confused because I can't reproduce the error with the code snippet from the issue #89 (and I see no difference of behavior between 2.1.0 tag and this bugfix branch)

That's interesting.. I can reproduce the issue on SDCC with IMAS-Python 2.1.0 (module load IMAS-Python). Could you try that as well?

I thought I did back then, but apparently not carefully enough, sorry for the delay.