Fix issue 622

[multimeric]

Closes #622.

---

Preview | Diff

[w3cbot]

gkellogg marked as non substantive for IPR from ash-nazg.

[w3cbot]

pchampin marked as substantive for IPR from ash-nazg.

[w3cbot]

This was discussed during the #json-ld meeting on 15 January 2025.

View the transcript

w3c/json-ld-api#623

<gb> Pull Request 623 Fix issue 622 (by multimeric)

bigbluehat: it looks like a white space reformat

gkellogg: it should have a change class label.

TallTed: it's more than just white space.

<niklasl> It says to fix w3c/json-ld-api#622 which is labelled class-3

<gb> Issue 622 Issue with Step 6 of `9.4.1 LoadDocumentCallback` (by multimeric) [spec:bug] [ErratumRaised] [class-3]

bigbluehat: good point niklasl.

gkellogg: this is class-2, it is not substantive.

TallTed: I disagree. The previous one said "X hold. reject", now it says "IF X holds, then reject".

gkellogg: ok, yes, it is class-3

pchampin: I'm a bit concerned we say this is both non-substantive + normative
… we don't not (yet) have an IPR commitment from this person
… but it seems we need it

gkellogg: we need to mark it back as substantive

pchampin: I'll see what I can do.
… I found it, marked it back as substantive, and sent a message to the person for them to commit to the IPR policy.

bigbluehat: ok, let's wait for their response.
… This change seems good, nobody seems to objet to it.

gkellogg: we need some tests for this.

<gkellogg> w3c/json-ld-api#622

<gb> Issue 622 Issue with Step 6 of `9.4.1 LoadDocumentCallback` (by multimeric) [spec:bug] [ErratumRaised] [class-3]

bigbluehat: anyone willing to write tests?

[crickets]

bigbluehat: ok, I'm marking it as "needs-test".

---

[pchampin]

thanks @multimeric for the IPR commitment.
@BigBlueHat @gkellogg we are good to move forward with this PR.

[gkellogg]

I added Candidate Correction markup and a log entry.

[pchampin]

There is a spurious change in the "changelog" section, which should be removed from this PR.
Otherwise, LGTM.

[pchampin]

Suggested change

	<li>2025-01-25: Correct some corner cases in transforming RDF Number Literals
	to JSON numbers when <a data-link-for="JsonLdOptions">useNativeTypes</a> is `true`,
	as described in <a href="#change_5">Candidate Correction 5</a>.</li>

I suspect this has leaked from another PR, right?
It should probably be removed from that one.

[multimeric]

This was added in 3b0a95c which I didn't make. Do you want me to fix it?

[gkellogg]

I added it to simplify a future merge, otherwise, it will lead to a conflict. If we do the PRs in an appropriate order, these will mesh together.