- From: Domenic Denicola <notifications@github.com>
- Date: Thu, 27 Oct 2022 22:16:12 -0700
- To: whatwg/webidl <webidl@noreply.github.com>
- Cc: Subscribed <subscribed@noreply.github.com>
- Message-ID: <whatwg/webidl/pull/1211/review/1159458895@github.com>
@domenic commented on this pull request.
> + [=deserialization steps=] preserve the additional information.
+
+<p class=note>These requirements mean that the inherited {{DOMException/code}} property of these
+interfaces will always return 0.
+
+<div class=example id=example-domexception-derived-interface>
+ The definition for a {{DOMException}} derived interface which carries along an additional
+ "hardware error code", for interfacing with a hypothetical Widget hardware device, could look
+ something like this:
+
+ <pre highlight=webidl>
+ [Exposed=Window]
+ interface WidgetError : DOMException {
+ constructor(optional DOMString message = "", WidgetErrorOptions options);
+
+ readonly attribute unsigned long long hardwareErrorCode;
I'm kind of envisioning this being some opaque number spit out by the widget hardware. I can see how maybe that's not a good pattern to have as our only example, as we'd arguably want to have a web platform API abstract over that sort of thing (like the SmartCard proposal is doing).
Using an enum, to me, gets into #1219, which I don't consider quite settled yet.
We should probably change this to be some number spit out by a protocol, instead of by a widget, since that's a bit more respectable and has precedent in both RTCError and WebTransportError.
--
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/webidl/pull/1211#discussion_r1007634047
You are receiving this because you are subscribed to this thread.
Message ID: <whatwg/webidl/pull/1211/review/1159458895@github.com>
Received on Friday, 28 October 2022 05:16:25 UTC