Re: [w3c/permissions] Model temporary permissions better (#86)

@jan-ivar

1. The `{state: 'prompt', access: obj}` strings I showed would be the return value from permissions.request() or any other request function in the platform, like a revamped `getCurrentLocation()`. The `access: obj` part would vary by permission. I think subsequent `query()`s within the same Realm would tend to return the same `state` value, although the UA is free to change that based on new information it receives.
1. The Safari store is *not* just per-origin, regardless of what we might want it to be. It has per-Realm state.

@alvestrand Could you elaborate a bit? I don't understand "'state: granted' as a new value of the state variable": 'granted' is an existing value for [PermissionState](https://w3c.github.io/permissions/#enumdef-permissionstate), which is the type of the [state](https://w3c.github.io/permissions/#permissionstatus) attribute. The current editors' draft does only [define the `state:` field in the dictionary](https://w3c.github.io/permissions/#enumdef-permissionstate) (not tuple), and leaves any other fields [up to the individual permission](https://w3c.github.io/permissions/#permission-result-type).

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/permissions/issues/86#issuecomment-215549499