Type II support for promise then is essential for hooking into existing promises' lifecycles and observing them without a massive amount of complexity.

[dead-claudia]

Mithril actively relies on it to observe when a promise chain stops, and removing this may break sites using it. (Some of the adjustments in there are because of an async/await change that spawned a bug report against us with multiple people complaining.)

This one is also isolated enough and simple enough to work around that I see value in it having a one-off exception, and I can't think of any other methods with such concerns.

[bakkot]

(Permalink for posterity.)

[bakkot]

Is this actually Type III? It doesn't look to me like it's overriding Symbol.species.

(Actually, it also looks to me like it's explicitly overwriting .then so that it wraps the result itself, rather than relying on the built-in .then to do it. But I'm less sure of that.)

[dead-claudia]

@bakkot It's overriding .then, not @@species. Probably should've specified it's more type II, not type III in hindsight, as that's more accurate. I'll change the title appropriately.

[codehag]

Type II is unlikely to be removed, we just have it there for completeness.