Re: Cookies: Integration with external specs.

[Mike West <[e-mail]>]

On Wed, Jan 20, 2016 at 5:25 AM, Mark Nottingham <[e-mail]> wrote:

>
> > On 19 Jan 2016, at 7:52 pm, Mike West <[e-mail]> wrote:
> >
> > On Tue, Jan 19, 2016 at 1:14 AM, Mark Nottingham <[e-mail]> wrote:
> > Hey Mike,
> >
> > On 18 Jan 2016, at 8:09 pm, Mike West <[e-mail]> wrote:
> > >
> > > While we have the cookies spec open, I think we should take a closer
> look at how that specification interacts with others. In particular, two
> things come to mind:
> > >
> > > * We should formalize the integration with Fetch (see step 11.1 of
> https://fetch.spec.whatwg.org/#http-network-or-cache-fetch and 9.3 of
> https://fetch.spec.whatwg.org/#http-network-fetch)
> >
> > What changes in the RFC would be necessary to do this?
> >
> > This would boil down to:
> >
> > * a light refactoring of https://tools.ietf.org/html/rfc6265#section-5.2
> away from "When the user agent receives a Set-Cookie header field in an
> HTTP response", towards something more like a "Process a `Set-Cookie`
> header" algorithm that Fetch can pass the header value into explicitly.
> >
> > * A thin shim on top of the "cookie string" algorithm that actually sets
> the header for a Request. This could live in Fetch, I suppose, but seems
> better positioned in the Cookie spec.
> >
> > I can put together a brief I-D spelling out these changes.
>
> Please.
>

Taking another look at the text, I don't think this is worth doing in
RFC6265bis. I've submitted https://github.com/whatwg/fetch/pull/221 to do
the work in the Fetch specification instead.

-mike