- From: Harald Alvestrand <harald@alvestrand.no>
- Date: Wed, 27 Jan 2016 12:11:23 +0100
- To: "public-media-capture@w3.org" <public-media-capture@w3.org>
- Cc: mlamouri@google.com, "public-webappsec@w3.org" <public-webappsec@w3.org>
Hi,
some time ago I made my proposal for MediaCapture permissions into a
pull request here:
https://github.com/w3c/permissions/pull/57
I've had some dialogue with Mounir about it afterwards (partly in the
PR), and I intend to make the following updates:
- Delete the "all-*" variants of permissions
- Add a new meaning to the descriptor: If there's no device in it, it
means "any"
- Add a new value for the descriptor: If the device is "*", it means "all".
Thus, you could say
navigator.permissions.query({name:'camera'}).then(function(result)) {
if (result.state == 'granted') {
// I know I can request "any camera" and get something, if any
cameras exist
navigator.permissions.query({name:'camera', deviceId='*'}).then(
if (result.state == 'granted') {
// I know I have access to all cameras that exist
navigator.permissions.query({name:'camera', deviceId='abcdef'}).then(
if (result.state == 'granted') {
// I know I have access to the specific camera with id 'abcdef'
Does this make sense to people?
(I'm not 100% sold on the usefulness of the '*' device ID myself...)
Since this crosses two WGs, perhaps comments on the PR itself is easiest.
Harald
Received on Wednesday, 27 January 2016 11:11:58 UTC