- From: Marijn Kruisselbrink <notifications@github.com>
- Date: Mon, 08 Aug 2016 10:33:41 -0700
- To: whatwg/fetch <fetch@noreply.github.com>
- Message-ID: <whatwg/fetch/issues/303/238311795@github.com>
Yeah, it would have that effect. Not sure if that is good or bad. For the notifications case it probably doesn't matter, since I don't think an installing worker is likely to be able to show notifications anyway. For other potential future uses I'm not sure what would be the desired effect though. It seems a bit weird to me that a fetch (indirectly) triggered by an installing worker would go through the older but still active worker, rather than through itself. But I agree that some fetches not going through the active worker is equally weird. But what would you expect with: SW1 at scope /foo, SW2 at scope /bar; SW1 creates a notification with an icon at /bar/icon.jpg (or even with multiple icons/resources, some at /bar/ and some at /foo/...). Should SW1 or SW2 intercept that icon? It seems a bit odd to me to treat the icon as its own top-level resource. And what if a document controlled by SW1 creates a notification associated with the registration for SW2? Should resources in that notification be fetched through SW1 (with or without the document as client) or SW2? --- 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-238311795
Received on Monday, 8 August 2016 19:50:53 UTC