Document overall lifecycle of media router mojo interfaces somewhere |
|||||
Issue descriptionIt's not particularly easy for someone who's not familiar to understand. There's logic to reconnect in case of a connection error, so MediaRouterMojoImpl ends up being a somewhat complicated state machine. For example, I couldn't figure out how pending events get drained when we reconnect, but I'm guessing the first thing the event page does is register itself with the MediaRouter, which causes pending callbacks to be run? Do we have this logic because extension renderer kills are something to be expected? Also, I was looking at DoStartObservingMediaRoutes and I wonder what happens if we register a route to start observing and the extension page crashes? When we restart it, how do we let the provider know to start observing those routes again?
,
Jul 27 2016
,
Apr 6 2018
We are going to update/revamp our public design doc in Q2 and it should include this information.
,
Apr 6 2018
,
Sep 24
,
Nov 7
|
|||||
►
Sign in to add a comment |
|||||
Comment 1 by mfo...@chromium.org
, Jul 25 2016Labels: -Pri-2 -M-54 Hotlist-CodeHealth Pri-3
Status: Untriaged (was: Available)