François Daoust

OK, thanks. If that file is useful, I'll try to re-create it in Webref, then. While you're all watching, feel free to have a look at https://github.com/speced/bikeshed/pull/2977, a proposal to...

There are two data levels in Webref: - the `main` branch contains raw data extracted from specs. That comes with no guarantee whatsoever. If a spec that defines CDDL contains...

Oh, and there's a third level I completely forgot to mention: - Updates are manually reviewed on a weekly basis (on average) to release npm packages: [`@webref/css`](https://www.npmjs.com/package/@webref/css), [`@webref/idl`](https://www.npmjs.com/package/@webref/idl), [`@webref/events`](https://www.npmjs.com/package/@webref/events). The...

> However, I’m still curious — if someone were to make a PR in, for example, the Permissions API that breaks the CDDL due to a conflict with the CDDL...

> The current plan to only add the local and remote indexes sounds good to me. Once concern will be that there are currently 38 types that are both local...

@OrKoN Back to the `all.cddl` file. I understand that's not needed to generate TypeScript types, but something that still bugs me is that the file in itself is not a...

That seems fine, @OrKoN. From a naming perspective, the current plan was for the crawler to create `webdriver-bidi-local.cddl`, `webdriver-bidi-remote.cddl`, and `webdriver-bidi.cddl` for the entire version. If that version is not...

@gsnedders said: > I don't think we need validation similar to what we have for WebIDL initially, just being able to find all the CDDL is itself already a win....

Webref now has [CDDL extracts](https://github.com/w3c/webref/tree/main/ed/cddl) of main specs that define CDDL: AT Driver, Permissions, Web Bluetooth, WebDriver BiDi. To be joined later on by a few other specs once the...

The ["run the scroll steps"](https://drafts.csswg.org/cssom-view-1/#document-run-the-scroll-steps) algorithm notest that the `scroll` and `scrollend` events bubble when target is a `Document`, which isn't correctly captured. Also, it seems strange that the algorithm...