More flexible way of passing avhrr_l1b_gaclac reader kwargs to pygac

[sfinkens]

Some reader arguments in avhrr_l1b_gaclac are directly passed to the corresponding pygac reader, and not used further in the satpy reader. To make the reader more flexible (and to prevent having to change satpy each time the reader signature in pygac changes), collect transitional arguments in **reader_kwargs and pass them to pygac. It doesn't obfuscate the call, because pygac will complain about invalid keyword arguments.

Furthermore, remove handling of derived metadata like midnight scanline and missing scanlines. They can easily be computed by users (via timestamps and qual flags) and so we have less work maintaining them.

• Tests added
• Tests passed
• Passes flake8 satpy

[coveralls]

Coverage decreased (-0.0003%) to 90.022% when pulling 1d23dd8 on sfinkens:pygac-metadata into 4b92a3c on pytroll:master.

[coveralls]

Coverage increased (+0.02%) to 90.04% when pulling e9997f7 on sfinkens:pygac-metadata into 4b92a3c on pytroll:master.

[codecov]

Codecov Report

Merging #1237 into master will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1237      +/-   ##
==========================================
+ Coverage   90.02%   90.03%   +0.01%     
==========================================
  Files         218      218              
  Lines       31259    31275      +16     
==========================================
+ Hits        28141    28160      +19     
+ Misses       3118     3115       -3     

Impacted Files	Coverage Δ
satpy/readers/avhrr_l1b_gaclac.py	95.03% <100.00%> (+1.69%)	⬆️
satpy/tests/reader_tests/test_avhrr_l1b_gaclac.py	99.58% <100.00%> (+0.01%)	⬆️
satpy/composites/__init__.py	80.70% <0.00%> (ø)
satpy/tests/reader_tests/test_agri_l1.py	100.00% <0.00%> (ø)
satpy/tests/test_composites.py	99.88% <0.00%> (+<0.01%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 4b92a3c...e9997f7. Read the comment docs.

[mraspaud]

Just a comment for now

[sfinkens]

@ninahakansson If you have time, I'd be pleased if you could review this PR

[ninahakansson]

Nice work! I like the more flexible use of kwargs. I tested this with
level1c4pps and it works fine.

I see that missing_scanlines are not any longer updated when data are sliced.
I am not sure if it could be a problem. But I agree that is is easier to not have to maintain
it inside satpy.

[sfinkens]

@ninahakansson Thanks for taking the time! Yes, that's true, at least for the moment. My plan is to remove the missing/midnight scanline attributes from pygac, too. In my opinion it makes more sense that the user computes them if needed. But we can certainly provide a method to compute that in pygac.

[mraspaud]

LGTM. Is this ready to merge @sfinkens ?

[sfinkens]

@mraspaud Yes please! 🙂