Ignore raw metadata when combining metadata

[sfinkens]

When combining metadata from multiple datasets, ignore the raw_metadata attribute. Usually these are nested dictionaries containing numpy arrays. You could compare them using np.testing.assert_equal, but

• raw metadata from another time/sensor will be different anyway
• this would slow down the code a little bit, because the raw metadata can be quite large
• we probably shouldn't use the numpy test module outside the unit tests

So I'd say it's not worth the effort.

• Closes Cannot load datasets of multiple SEVIRI native files  #1656
• Tests added

[codecov]

Codecov Report

Merging #1663 (202c6fd) into main (6df2f10) will increase coverage by 0.00%.
The diff coverage is 100.00%.

❗ Current head 202c6fd differs from pull request most recent head 5f747c4. Consider uploading reports for the commit 5f747c4 to get more accurate results

@@           Coverage Diff           @@
##             main    #1663   +/-   ##
=======================================
  Coverage   92.65%   92.65%           
=======================================
  Files         258      258           
  Lines       37948    37952    +4     
=======================================
+ Hits        35161    35165    +4     
  Misses       2787     2787           

Flag	Coverage Δ
behaviourtests	4.83% <0.00%> (-0.01%)	⬇️
unittests	92.79% <100.00%> (+<0.01%)	⬆️

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

Impacted Files	Coverage Δ
satpy/tests/test_dataset.py	100.00% <ø> (ø)
satpy/dataset/metadata.py	98.73% <100.00%> (+0.06%)	⬆️

---

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 6df2f10...5f747c4. Read the comment docs.

[djhoese]

Thanks for working on this. I had a couple suggestions.

[djhoese]

LGTM. Thanks!

[djhoese]

Why did travis and appveyor suddenly show up again in this PR? Did you have an old master/main branch that you based this branch off of?

[sfinkens]

Hmm, that is strange. The branch is based on 6df2f10 which was a couple commits before master was renamed to main (did that in my fork, too)

[ameraner]

LGTM, thanks a lot for your help fixing this!