Re: [whatwg/fetch] Odd format for fetch callbacks (#536)

> Actually the DOM fetch method calls the second one from "in-parallel" steps(see step 9).

To be clear, I think that's problematic, as it means the CSP policy gets enforced from in parallel and it isn't clear at what point it is copied to make that deterministic. (In part this is a separate issue, but I find it useful to think about as it influences what kind of algorithms we should be offering to standards.)

---

I guess since the callbacks take a set of steps you could indeed give them any kind of variables they might need to take hold of. It does seem like that approach might work quite well.

I'd love it if one or more of @yutakahirano, @jyasskin, @jakearchibald, @domenic, and maybe @jungkees could take a look as well.

(I suspect that for most standards we want to provide a wrapper algorithm that preserves the status quo more or less, whereby you invoke "fetch with tasks" or some such and it continues queuing tasks for the callback steps.)

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/fetch/issues/536#issuecomment-640454679