Add Clipboard Feature Policy

[dway123]

Closes #106

The following tasks have been completed:

• Confirmed there are no ReSpec/BikeShed errors or warnings.
• Modified Web platform tests

Implementation commitment:

• WebKit (https://bugs.webkit.org/show_bug.cgi?id=)
• Chromium (https://crrev.com/c/2164248)
• Gecko (https://bugzilla.mozilla.org/show_bug.cgi?id=1646270)

---

Preview | Diff

[marcoscaceres]

again, it's kinda annoying that we ended up with "clipboard-read" and "clipboard-write" as the feature name :( ... would have been nice to have a single name, as we are again potentially setting a bad precedence.

[dway123]

Thanks for the review! I've updated the change a bit now per the comments.

Yeah, I agree that a single feature name would have been nice as well, but am just trying to match the clipboard spec's permission names with the feature policy, as feature policy and permissions are merging since this change.

[marcoscaceres]

Filed bug on the Gecko side too.

[marcoscaceres]

This current algorithm currently continues with the rejected promise, instead of returning it. Let's do:

Suggested change

	1. Let |p| be a new [=Promise=].
	1. If the [=current settings object's=] responsible document is not
	[=allowed to use=] the "clipboard-write" feature, then reject |p|
	with a "NotAllowedError" DOMException.
	1. If the [=current settings object's=] responsible document is not
	[=allowed to use=] the "clipboard-write" feature, then
	return a promise rejected with a "NotAllowedError" DOMException.
	1. Let |p| be a new [=Promise=].

[marcoscaceres]

Just some nits... I think you can link [=responsible document=] and [=a promise rejected with=] ... please confirm by looking up the terms at https://respec.org/xref/ ... you can probably link {{DOMException}} too.

[marcoscaceres]

Same as above... immediately return the rejected promise, otherwise continue in parallel.

[marcoscaceres]

As above here too...

[marcoscaceres]

Same... 😊

[marcoscaceres]

Super close... what might be interesting now is to rebase this given the transient activation text is in the spec... we might be able to also do the transient activation checks early.

[saschanaz]

again, it's kinda annoying that we ended up with "clipboard-read" and "clipboard-write" as the feature name :(

Are we too late to rename them? It seems currently Chrome has the only implementation, should we discuss about renaming? (I think I saw somewhere that WebKit has no intent to implement them. Edit: Yes, #101 (comment))

[marcoscaceres]

Yes, if only Firefox and Chrome are supporting these, we should discuss renaming them... but we would need to commit to implementing them too.

[saschanaz]

Firefox has no implementation, only Chrome does.

[marcoscaceres]

Firefox has no implementation, only Chrome does.

Understood, but we should figure out if we will implement it in the future (and figure out if what's in the spec is palatable to us, and remove what Gecko is not willing to support to match reality).

[wandyezj]

What is really annoying is that browsers have gone ahead with their own implementation for a critical feature. This ends up forcing developers to use commands to read and write to the clipboard to work around this nonsense.