- From: Marcos Cáceres <notifications@github.com>
- Date: Mon, 24 Oct 2016 05:33:38 -0700
- To: w3c/manifest <manifest@noreply.github.com>
- Message-ID: <w3c/manifest/pull/506/review/5437954@github.com>
marcoscaceres commented on this pull request.
> + didPrompt: false,
+ userChoice: null,
+ userChoiceResolver: null, // Implicit in spec
+ };
+
+ internal.userChoice = new Promise((resolve) => {
+ if (eventInit && "userChoice" in eventInit) {
+ return resolve(eventInit.userChoice);
+ }
+ internal.userChoiceResolver = resolve;
+ });
+ internalSlots.set(this, internal);
+ }
+
+ prompt() {
+ if (this.isTrusted === false) {
Well, it's as functional as any other user-constructed event that relies on trust. You are correct that it's crippled - but that's again by design: you can still construct these events and fire them, but it's just "prompt()" that doesn't do anything - but that will be exactly the same for the real ones when constructed by developers. In that sense, what is currently implemented is correct per-spec.
Note that I could, theoretically, take BeforeInstallPromptEvent.js and plug it into Gecko, because we support JS-based IDL bindings. So, in such an implementation I would get a "trusted" event.
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/w3c/manifest/pull/506
Received on Monday, 24 October 2016 12:34:15 UTC