- From: Darien Maillet Valentine <notifications@github.com>
- Date: Wed, 16 Aug 2023 08:41:13 -0700
- To: whatwg/webidl <webidl@noreply.github.com>
- Cc: Subscribed <subscribed@noreply.github.com>
Received on Wednesday, 16 August 2023 15:41:19 UTC
_This comment only intends to make a design decision implied by this PR explicit to other readers._ The [LTNOAN] annotation causes only _nullable_ types with the annotated callback function as their inner type to have the problematically expanded value space — and more specifically, only when used as an attribute’s type. In every other context, an [LTNOAN]-annotated callback function behaves exactly the same as any other. This PR will leave all [LTNOAN]-annotated callback functions indistinguishable from dictionary-likes even in contexts where [LTNOAN] is otherwise ignored/meaningless. That seems like a reasonable call: - It’s much simpler than trying to account for all those quirky conditions - It’s usually okay (or even desirable) to leave _legacy_ features in the dust when introducing new features Again, this comment’s just an “FYI” thing in case other folks _do_ see a problem with this approach — as far as I can tell, it works. -- Reply to this email directly or view it on GitHub: https://github.com/whatwg/webidl/pull/1353#issuecomment-1680849505 You are receiving this because you are subscribed to this thread. Message ID: <whatwg/webidl/pull/1353/c1680849505@github.com>
Received on Wednesday, 16 August 2023 15:41:19 UTC