SPDX rdf tests pass automatically

[armintaenzertng]

Description

The SPDX rdf tests currently pass automatically. This is due to this line in the code, which basically sets expected == result per default.
The bug can be fixed by removing the linked line.

As almost all spdx tests are skipped via the @pytest.mark.scanslow annotation, this does not affect most of the tests.
However, if you remove these annotations they will fail with the above fix.

When upgrading the spdx-tools to the new 0.8 version, these test have to be slightly adapted, anyway. I can offer to do this while working on the library upgrade.

How To Reproduce

Remove this line in the code and execute for example this test (you have to remove the annotation first).

System configuration

Ubuntu 22.04.2 LTS, Scancode 32.0.5rc3

[armintaenzertng]

I'm not sure about the other part of the if clause, which will also pass automatically due to this line. But maybe that is the intended behavior in that case?

[pombredanne]

Great catch!
So the behaviour should be:

1. if the test fixture regeneration environment variable REGEN_TEST_FIXTURES is set then the expected results file should be update with the latest results content.
2. otherwise, the test should run and the results checked against pre-computed expected results

Therefore the whole block at

scancode-toolkit/tests/formattedcode/test_output_spdx.py

Lines 128 to 131 in a15174f

	with io.open(expected_file, encoding='utf-8') as i:
	expected = json.load(i)
	expected = load_and_clean_rdf(result_file)

should be replaced by this:

expected = load_and_clean_rdf(expected_file)

[pombredanne]

Can I interest you in a patch? 👼

[armintaenzertng]

Can I interest you in a patch?

Yes, most certainly! :)
Can I interest you in our new spdx-tools version 0.8? :D

[armintaenzertng]

@pombredanne, just in case you missed it, I opened a PR to fix this in #3451.