Re: HPACK, Dynamic Table Size Update

On 9 June 2015 at 13:57, Pavel Rappo <pavel.rappo@gmail.com> wrote:
> To me it seems like it _only_ applies to solicited updates.

That is my understanding too (though it might not be well-tested; you
can check with the test suites here:
https://github.com/http2jp/hpack-test-case)

> Just curious, what was the rationale behind binding size updates to a specific
> location in a header block (for the solicited case)?

The rule here is that the change had to be immediate.  That is, once
you acknowledge the setting at the HTTP/2 framing layer, you can't
delay enacting the change within HPACK.

> 2. As I understand a header block cannot consist of only size updates because
> "size update" is not a header field representation. So size updates (if any) has
> to go with some header fields. Am I right?

Well, technically a header block can be empty, but only for trailers.
All the other blocks have mandatory header fields.  It's an odd case,
and (again) might not be well-tested, so I wouldn't do it.  But it
isn't technically disallowed.

Received on Tuesday, 9 June 2015 22:13:26 UTC