Re: [webidl] Add a [SecureContext] operator attribute (#65)

> It does, but I would prefer we do it after the "a security check" bit.

As long as the check happens somewhere around here, I think we're fine. The exact ordering isn't something I have strong feelings about, so I'm happy to move it around. Moved under "a security check" in the latest patch.

> That's probably fine as long as it's explicitly marked informative.

Wrapped in a `<div class="note">` in the latest patch.

> whether two pages that are same-effectives-script-origin can ever differ in terms of whether they're secure contexts.

Since the notion walks the ancestor tree, yes, two same-origin contexts can have different security states. I don't think that will end up being a terribly difficult check to do in Blink's bindings, and I think it's the behavior we've already implemented for restrictions on things like WebCrypto and WebRTC.

---
Reply to this email directly or view it on GitHub:
https://github.com/heycam/webidl/pull/65#issuecomment-149260729