Krzysztof Kotowicz

Friendly ping, cc @gaearon. This would probably need rebasing, but the open question is whether we should go with: * the changes here - These practically unblock using react-DOM with...

> It's ok to drop IE9 by this point. If you have a fix that only breaks something in IE9 without adding extra logic let's merge it. Just done in...

That's https://github.com/whatwg/html/issues/4651. We don't have a client passed to the algorithm, but that also fails in CSP before that happens.

This seems to be resolved in the spec via https://github.com/w3c/trusted-types/pull/557, and in WPT via https://github.com/web-platform-tests/wpt/pull/48769. @fred-wang, do you agree this can be closed now?

Realistically we only have two choices, either ignore it (as `baseVal` is a property of `SVGAnimatedString` and not an element), or use a meta syntax like `href.baseVal` for the property...

The test seems to be added already, the finals bit in https://github.com/web-platform-tests/wpt/commit/4db7dc692be3aca644252623d69d5db5113541f7. @mbrodesser-Igalia, can this be closed now?

This is intentional. The default policy only works if there's a `require-trusted-types-for` directive. This is such that all trusted types related enforcement is controlled through the directive.

Yeah, we can revert this.

Does that not require defining the stringification in https://w3c.github.io/trusted-types/dist/spec/#process-value-with-a-default-policy-algorithm anyway? This is the only callsite that does not already pass a string.

Re: defining stringification for TrustedType objects, the spec [uses](https://w3c.github.io/trusted-types/dist/spec/#trustedhtml-stringification-behavior) the [WebIDL stringifiers](https://webidl.spec.whatwg.org/#idl-stringifiers) - they are not linked to WebIDL definitions, but that syntax seems to be [recommended by Bikeshed](https://speced.github.io/bikeshed/#linking-to-stringifiers). Is...