Open aucampia opened 2 years ago
for info, lists of known bugs at https://github.com/RDFLib/rdflib/blob/6f2c11cd2c549d6410f9a1c948ab3a8dbf77ca00/test/jsonld/test_onedotone.py#L27 and https://github.com/RDFLib/rdflib/blob/6f2c11cd2c549d6410f9a1c948ab3a8dbf77ca00/test/jsonld/test_testsuite.py#L22 - this latter also references "unsupported tests" (whatever they are).
Am I understanding this correctly by saying that the current JSON-LD test runner is tied to rdflib/plugins/parsers/jsonld.py
?
Instead of calling the JSON-LD parser through a Graph or a Dataset, it calls the to_rdf()
directly.
Am I understanding this correctly by saying that the current JSON-LD test runner is tied to
rdflib/plugins/parsers/jsonld.py
?
I'm actually not sure about this at all, all I'm fairly confident about is that we don't run the whole test suite which is something which we should be doing with priority. But I also agree test suite should run via the main RDFLib interfaces, not via plugin or JSON-LD specific interfaces, at least where possible, there is more to JSON-LD though which complicates things, but the highest priority is that we run and pass the test suite for things that is covered by the main RDFLib APIs.
We should be running all JSON-LD 1.1 tests For JSON-LD 1.0 test see #1832.