- From: Kinuko Yasuda <notifications@github.com>
- Date: Tue, 07 Apr 2015 01:07:57 -0700
- To: slightlyoff/ServiceWorker <ServiceWorker@noreply.github.com>
Received on Tuesday, 7 April 2015 08:08:21 UTC
I agree that there could probably be a situation where one wants to explicitly extend the lifetime of an event (and hence ServiceWorker) up to a certain point after it returns a response, assuming that in general we want ServiceWorker to return a response as early as possible for performance reasons. Reg: TCP timeout-- it's not clear if we need to tie the lifetime of SW to the completion of async network/cache operation. It looks we could say that async network/cache operation that is triggered by a ServiceWorker will NOT be cancelled even after the ServiceWorker is killed (by UA), like we don't for Beacon API on page unload. (Currently the behavior of async operations upon SW termination doesn't seem to be really spec'ed?) --- Reply to this email directly or view it on GitHub: https://github.com/slightlyoff/ServiceWorker/issues/584#issuecomment-90448085
Received on Tuesday, 7 April 2015 08:08:21 UTC