6.2.0: pytest is failing

[kloczek]

I'm packaging your module as an rpm package so I'm using the typical PEP517 based build, install and test cycle used on building packages from non-root account.

• python3 -sBm build -w --no-isolation
• because I'm calling build with --no-isolation I'm using during all processes only locally installed modules
• install .whl file in </install/prefix>
• run pytest with PYTHONPATH pointing to sitearch and sitelib inside </install/prefix>

[kloczek]

Here is pytest output
python-rdflib-pytest-log.txt

[aucampia]

@kloczek this is a known issue: #1847

To work around it, don't run tests with lxml installed.

[aucampia]

I will update our CI to run with lxml at least on some versions of python and then mark the failing tests with xfail.

[kloczek]

@kloczek this is a known issue: #1847

To work around it, don't run tests with lxml installed.

I'm not sure do I understand that ticket correctly.
Does it mean that it is something wrong with lxml or rdflib needs to be updated? 🤔

[aucampia]

RDFLib has a bug which is tracked in #1847 - and because of this bug, using lxml with XML based SPARQL results fail.

We should fix the bug, but we should also mark the tests in our test suite that are known to fail under some conditions with xfail under those conditions, so that the test suite still passes.

For example:

rdflib/test/test_w3c_spec/test_n3_w3c.py

Lines 117 to 119 in d263009

	if os.name == "nt":
	for test in ["turtle-subm-15", "turtle-subm-16"]:
	EXPECTED_FAILURES[test] = "Issue with nt parser and line endings on windows"

This test is known to fail on windows, so we mark it with xfail on windows.

Similarly, with lxml installed, the failing tests here should have an xfail marker.

[kloczek]

Thank you very much for the explanation 😄