Skip to content

Update extensions registry - #2642

Merged
weegeekps merged 1 commit into
KhronosGroup:mainfrom
lexaknyazev:ext-upd
Sep 3, 2026
Merged

Update extensions registry#2642
weegeekps merged 1 commit into
KhronosGroup:mainfrom
lexaknyazev:ext-upd

Conversation

@lexaknyazev

Copy link
Copy Markdown
Member

@weegeekps could we add anything to the line 536?

@weegeekps

Copy link
Copy Markdown
Contributor

@lexaknyazev Yes, maybe these?

Esri may have something to share for ArcGIS too, but I am unsure if any of their implementations are open source. Not sure what link they would want to use.

@lexaknyazev

Copy link
Copy Markdown
Member Author

We don't need to list all known implementations, just something to replace the placeholder.

@javagl

javagl commented Sep 2, 2026

Copy link
Copy Markdown
Contributor

We had some discussions about this, and right now, I don't want to engage in arguments about what "what something being supported" means (and if so, that could/should happen at #2542). But people might want to try and load #2562 and expect the implementation to load them, and if this is not possible then it should be made clear. (I don't even know what kind of error message to expect there right now, but I'd place my bets on a DeveloperError being thrown)

@lexaknyazev

Copy link
Copy Markdown
Member Author

@javagl The question for this PR is much simpler: would you prefer JSplat to be mentioned in the extension or not?

@javagl

javagl commented Sep 3, 2026

Copy link
Copy Markdown
Contributor

Prefer - compared to what?

Many extensions don't list "Known Implementations". Some just list "References", linking to papers, or GitHub permalinks to readers/writers (but from a quick sample, the latter is not used for the ratified ones). I'm still convinced that things like Draco don't even have two independent implementations.

In any case, this list could only be a "snapshot", as it will not be updated, and it could only be a formality, because it doesn't point to the relevant place in the code or make claims about whether the implementation is a reader, writer, or renderer.

So if a tool that can write the extension (but not read it) and a renderer that can render the SPZ extension (but not the base extension) count as "two implementations of the (base?) extension", in the spirit of the ratification conditions of the "Khronos Operational Guidelines", then that may be enough. JSplat can read, write, and render the base extension (but no other aspects of glTF, because it's focussed on splats). I frankly don't care whether or not it is included.

Bottom line: I have no strong preference here.

@lexaknyazev

Copy link
Copy Markdown
Member Author

@weegeekps PTAL.

@weegeekps weegeekps left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM. 🚢

@weegeekps
weegeekps merged commit 81762cc into KhronosGroup:main Sep 3, 2026
2 checks passed
@lexaknyazev
lexaknyazev deleted the ext-upd branch September 3, 2026 13:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants