Spec text for the identity reference space #520
Conversation
| 1. Let |options| be the {{XRReferenceSpaceOptions}} passed to {{requestReferenceSpace()}}. | ||
| 1. Let |type| be set to |options| {{XRReferenceSpaceOptions/type}}. | ||
| 1. Let |referenceSpace| be set to <code>null</code>. | ||
| 1. If |type| is {{identity}} let |referenceSpace| be a new {{XRReferenceSpace}}. |
There was a problem hiding this comment.
Don't we also need to error if a subtype is present?
There was a problem hiding this comment.
We don't really need to, in my opinion. If you passed in any other non-applicable values in the dictionary (ie: { type: 'identity', foo: 'bar' }) then we just ignore it, so that feels like it could carry over to this as well? Then again, if you passed a subtype that wasn't valid the IDL would cause it to naturally error whether or not the type uses it, so maybe being more rigorous about it isn't a bad idea? IDK.
There was a problem hiding this comment.
My general opinion is that we should do what's consistent with the rest of the web on this one. Got an API ergonomics person who wants to give us guidance one way or another?
toji
force-pushed
the
spec-identity
branch
2 times, most recently
from
9a0f788 to
c01803a
Compare
Fixes #482
Adding the
"identity"reference space type. Not includingoriginOffsetat this point to keep PRs focused and easy to review.