Fix set semantics bug with other store interface

[maartyman]

@rubensworks is this what you had in mind?

[coveralls]

Pull Request Test Coverage Report for Build 7609127000

• 0 of 0 changed or added relevant lines in 0 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage remained the same at 100.0%

---

Totals
Change from base Build 7552705868:	0.0%
Covered Lines:	59
Relevant Lines:	59

---

💛 - Coveralls

[rubensworks]

Yes, this looks perfect to me!

@surilindur could you try this out in link-traversal master and your void branch?

[rubensworks]

Suggested change

	subject: RDF.Term | null,
	predicate: RDF.Term | null,
	object: RDF.Term | null,
	graph: RDF.Term | null,
	subject?: RDF.Term | null,
	predicate?: RDF.Term | null,
	object?: RDF.Term | null,
	graph?: RDF.Term | null,

[maartyman]

n3 store doesn't support undefined: https://github.com/DefinitelyTyped/DefinitelyTyped/blob/4c017214b1c3fcfd5bb1637035f806b06c6b76d3/types/n3/index.d.ts#L295 the OTerm can only be a RDF.Term, string or null.

[rubensworks]

It actually does support undefined: https://github.com/rdfjs/N3.js/blob/main/src/N3Store.js#L445-L447
But it appears the typings are not fully correct.

So yeah, let's keep it as-is for now. We can modify later once the N3 typings would get fixed.

[maartyman]

I assume this is support for when OTerm is null but the logic is the same for undefined.

[rubensworks]

Indeed, but since null is not the same as undefined, tsc will complain about it.

[surilindur]

This seems to produce the same results as the current 1.1.2 version in link traversal master branch, so it definitely does not seem to break that this time. I am also fairly sure I tested the correct versions of everything this time. Edit: also the tests in link traversal pass.

I will try to test with the VoID stuff soon, either today or tomorrow.

[surilindur]

After testing the version from this branch with the VoID-related work, it also seems to consistently produce the same results every time, with no spurious duplicates, since I cannot seem to be able to reproduce the cases where it was producing random numbers of results or exactly double the expected.

[rubensworks]

Excellent, thanks for checking @surilindur, and thanks for the fix @maartyman!