Towards default nominal angles

[kmuehlbauer]

• Possible successor of ENH: extract nominal ray angle by default for ODIM backend #383
• Tests added
• Changes are documented in history.md

[kmuehlbauer]

@egouden I've taken your #383 and tweaked it toward my suggestion.

When you find the time, please have a look if this would also work from your perspective.

[codecov]

Codecov Report

❌ Patch coverage is 89.09091% with 6 lines in your changes missing coverage. Please review.
✅ Project coverage is 94.19%. Comparing base (79ba495) to head (de781a5).

Files with missing lines	Patch %	Lines
xradar/io/backends/odim.py	89.09%	6 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #386      +/-   ##
==========================================
- Coverage   94.20%   94.19%   -0.01%     
==========================================
  Files          29       29              
  Lines        6417     6461      +44     
==========================================
+ Hits         6045     6086      +41     
- Misses        372      375       +3     

Flag	Coverage Δ
notebooktests	0.00% <0.00%> (ø)
unittests	94.19% <89.09%> (-0.01%)	⬇️

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

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[egouden]

Thanks @kmuehlbauer for handling this issue.

This is a more elaborated solution. It remains quite simple. I like it.

Is it the plan to tackle each format separately? I think for rainbow the nominal angle is included in the metadata.

[kmuehlbauer]

@egouden The only thing I'm still not quite sure:

If we have a missing and a duplicate ray for some reason, like 0, 0, 1,..., 357, 359, it would create the nominal angles 0, 1, ..., 358, 359. Finally this would return an array which is one off for most of the sweep. I'll ponder about this a bit more, maybe we can get this right, too.

About the nominal angles: if formats support that, we should implement it. Otherwise we still could get nominal angles by auto extraction/reindex_dict.

Would be nice to have conforming behaviour throughout the readers, though.