Re: HPACK: pseudo-headers ordering

On 6 August 2014 03:31, Kaoru Maeda <> wrote:

> Does the word "MUST" states that HPACK should bring pseudo-
> headers to the beginning?

The implication here by having this requirement in HTTP/2 and not HPACK, is
that the HTTP mapping maintains this.  Though this is largely a division of
responsibilities thing that we don't usually worry about here, you are free
to structure your code as you choose.  If I were building HPACK, I wouldn't
bother even looking at the header field names.

Received on Wednesday, 6 August 2014 18:37:51 UTC