Add a testing utility for faking reading

[mraspaud]

This PR adds a testing utility to satpy for faking reading.

Useful for testing programs that use satpy for reading with actually needing to read anything.
Can be used like this:

        scene_dict = {channel: somedata}
        with fake_satpy_reading(scene_dict):
            scene = Scene(input_files, reader="dummy_reader")
            scene.load([channel])

and the scene will be loaded with the contents of scene_dict.

• Closes #xxxx
• Tests added
• Fully documented
• Add your name to AUTHORS.md if not there already

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.98%. Comparing base (1832175) to head (f9dba55).
Report is 467 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2840      +/-   ##
==========================================
+ Coverage   95.95%   95.98%   +0.03%     
==========================================
  Files         366      368       +2     
  Lines       53615    53798     +183     
==========================================
+ Hits        51446    51639     +193     
+ Misses       2169     2159      -10     

Flag	Coverage Δ
behaviourtests	4.03% <0.00%> (-0.02%)	⬇️
unittests	96.08% <100.00%> (+0.03%)	⬆️

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

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

[djhoese]

Interesting. Makes sense. Some concerns:

1. I realize this is for users and their applications, but should this be in satpy/tests/ instead?
2. Could we have a new section in the documentation (may require a whole new page) regarding "Testing your Satpy code"?
3. Users may want to debug their custom composites, both configs and actual execution of custom composite python code. Is there a way to allow for that? I suppose this would require mocking the lower-level Scene readers dictionary and distinguishing between the users scene_dict being reader datasets versus composite datasets. Maybe not worth it.

[mraspaud]

1. I'm following numpy here, as they have np.testing. I would also add that tests for me are rather internal/private, while testing is intended to be public api.

[djhoese]

Sounds good.

[gerritholl]

Users may want to debug their custom composites, both configs and actual execution of custom composite python code. Is there a way to allow for that? I suppose this would require mocking the lower-level Scene readers dictionary and distinguishing between the users scene_dict being reader datasets versus composite datasets. Maybe not worth it.

The final three tests at https://github.com/pytroll/satpy/blob/main/satpy/tests/modifier_tests/test_parallax.py test loading a composite via the Scene.load interface, involving a fake configuration file and mocking satpy.composites.config_loader.config_search_paths.

Maybe those tests could use this utility instead (not sure).

[coveralls]

Pull Request Test Coverage Report for Build 10058493408

Warning: This coverage report may be inaccurate.

This pull request's base commit is no longer the HEAD commit of its target branch. This means it includes changes from outside the original pull request, including, potentially, unrelated coverage changes.

• For more information on this, see Tracking coverage changes with pull request builds.
• To avoid this issue with future PRs, see these Recommended CI Configurations.
• For a quick fix, rebase this PR at GitHub. Your next report should be accurate.

Details

• 34 of 34 (100.0%) changed or added relevant lines in 2 files are covered.
• 7 unchanged lines in 2 files lost coverage.
• Overall coverage increased (+0.02%) to 96.086%

---

Files with Coverage Reduction	New Missed Lines	%
satpy/tests/reader_tests/test_ami_l1b.py	3	98.17%
satpy/readers/ami_l1b.py	4	97.32%

Totals
Change from base Build 10006768762:	0.02%
Covered Lines:	51868
Relevant Lines:	53981

---

💛 - Coveralls

[mraspaud]

2. addressed here https://satpy--2840.org.readthedocs.build/en/2840/dev_guide/testing.html

[djhoese]

This isn't a hard requirement/change for this to be merged, but this wording of "faking reading" seems not as descriptive as I'd like. On one hand I would expect the high-level description to be something like "To allow Scene creation and loading fake datasets without reading files", but that's kind of long. My main point being that "For faking reading" doesn't seem to fully describe what is possible, or rather, why you'd use this functionality.

[mraspaud]

I like your wording, so I'll put it in here. The docstring of that function also has an example, so that part is quite clear I think.