James Browning

Inner calls may still need to be transformed though e.g. ```javascript function square() ... function foo() { console.log(square(10)) } ``` would need to become ```javascript ... function foo() { const...

I was not aware of `enter()/exit()` that solves basically everything I need. I'll still leave this issue open but that info gives a pretty simple solution for traverseUp, it'd essentially...

Oh also `traverseUp` would need to add `path.skip()` after each `exit()` implicity (so that it doesn't retraverse the node when replaced). As for the handbook it'd be nice to have...

> Wouldn't that mean that `signal.aborted !== signal.aborted` sometimes? Yes. Although once `true`, always `true`. > I don't think we want that. Is there any particular reason? Web APIs that...

> That isn't true? `aborted` can only be observed on the main thread, and if author JavaScript is running on the main thread, then web API browser code cannot be....

Just to clarify how I'd imagine this working, if `AbortSignal` were implemented in js I would imagine doing something like this. I've added an explicit `synchronizeAborted` method to this example,...

With the new `abortSignal.reason` synchronous observation doesn't really make sense anymore as one also needs to send the reason across the thread which is considerably less simple than the setting...

> That seems reasonable to me and is compatible with what we are planning to do for Fetch (see [whatwg/fetch#1343](https://github.com/whatwg/fetch/pull/1343)). (The current plan there is to serialize/deserialize abort reason and...

> Just an update on this... we recently just landed early support for this in Node.js. When sending an `AbortSignal` over postMessage, a new `AbortSignal` is created on the receiving...

> That is exactly what I was envisioning - a small if (this.shadowRoot) block that just hooks things up if the shadowRoot already exists, and the actual construction code if...