Closed zecakeh closed 2 months ago
This isn't really rendering right in the preview - it's referring to a OneTimeKeys
type which is then not defined. Is that something that is fixed by one of the other PRs in flight?
I am wondering if the oneOf: [string, KeyObject] is right, since the section about one-time and fallback keys clearly says that only signed_curve25519 is used in this case.
Yeah. I think the problem is that, previously, curve25519
was used here, where the type is just string
. I think it's fair to say that nobody should be using unsigned curve25519
keys here, but maybe that is a separate change?
This isn't really rendering right in the preview - it's referring to a OneTimeKeys type which is then not defined. Is that something that is fixed by one of the other PRs in flight?
Yes, that is fixed by #1801. Or an alternative is to remove the title
.
This was commented prior to the port to OpenAPI 3.1 for technical reasons (#1127). Now we can use it just fine.
I am wondering if the
oneOf: [string, KeyObject]
is right, since the section about one-time and fallback keys clearly says that onlysigned_curve25519
is used in this case.This adds the
KeyObject
definition in several places:Preview: https://pr1800--matrix-spec-previews.netlify.app