Closed alex-chromium closed 4 years ago
I think we should only generate the "license-release" message in remove() in all cases for simplicity and consistency. Thoughts?
Yes, the load()
method does not generate such a message, so the section on Clear Key behavior of "persistent-usage-record"
is in error. I think we should just drop mention of load()
from that line.
Thanks! I'll upload a pull request.
In section 5 (https://w3c.github.io/encrypted-media/#idl-def-MediaKeySessionType.persistent-usage-record) for persistent-usage-record it says
A message of type "license-release" containing the record of key usage will be generated each time remove() is called
with no mention of theload()
method.Further down, in section 9.1.2 (https://w3c.github.io/encrypted-media/#clear-key-behavior), the specs mentions a message containing the record of key usage as part of the
load()
algorithm.For sessions of type "persistent-usage-record", in the remove() and load() algorithms, the message reflecting the object's record of key usage is a JSON object encoded in UTF-8 as described in License Release Format.
This feel like a spec bug, should the reference to
load()
be remove from the spec in section 9.1.2?