- From: Ben Kelly <notifications@github.com>
- Date: Tue, 09 Aug 2016 07:44:37 -0700
- To: whatwg/fetch <fetch@noreply.github.com>
Received on Tuesday, 9 August 2016 14:45:33 UTC
At the meeting I wanted a controlling client. Today we have two mechanisms to keep a service worker in the `activated` state: 1) Clients are being controlled 2) The SW thread is actively handling events or being held alive by extension promises (this is newish) If we say that the notification requests should just be "intercepted by the SW that made the request", then we run the risk of the SW going redundant before the FetchEvent is fired. What guarantees that won't happen? I don't think either of the two cases above are guaranteed to be in effect here. It would be breaking for a redundant worker to handle a FetchEvent because its state will have been completely stomped on by the updated SW version that took its place. --- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/whatwg/fetch/issues/303#issuecomment-238575904
Received on Tuesday, 9 August 2016 14:45:33 UTC