Re: [whatwg/fetch] Use case for Headers getAll (#973)

What is the conclusion here? I'd be happy to put in the work to get this landed, but I am not sure exactly what to do. My current understanding:

- There is no strict objections to the proposed behaviour of special handling of `set-cookie` in `Headers` objects.
- Implementers are worried about the additional complexity of supporting `set-cookie` headers in the Cache API as the internal data structure might have to change.
- The above issue could be mitigated by not allowing `set-cookie` headers to be present in `Request` (like is already the case in `Response`) in browsers, because then this header would never be able to make it into a `Request`.
- The above seems fine for all parties, but we need to figure out if this change would break users. The inclination is "no", but it needs to be verified.

That suggests that the next step is to measure usage of `set-cookie` headers in `Headers` objects with a `"request"` header guard. I expect this would mean adding a use counter in one of Chromium or Gecko. I have no idea how to do this: I tried doing it for Gecko just now, but I can not figure it out. I will also try for Chromium now. I'd appreciate any help on this front.

-- 
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/fetch/issues/973#issuecomment-1025227167
You are receiving this because you are subscribed to this thread.

Message ID: <whatwg/fetch/issues/973/1025227167@github.com>