- From: Martin Thomson <martin.thomson@gmail.com>
- Date: 2013年8月13日 00:58:38 +0100
- To: Jeff Pinner <jpinner@twitter.com>
- Cc: James M Snell <jasnell@gmail.com>, HTTP Working Group <ietf-http-wg@w3.org>
- Message-ID: <CABkgnnUy4qmx6pNJq+6W4=a0R180rgVVNbDiVxiFyWVd5pWVZQ@mail.gmail.com>
On 13 August 2013 00:38, Jeff Pinner <jpinner@twitter.com> wrote: > I think it might be better to move the logic off rather than add flags for > all frames that might be continued. This keeps the mechanism as simple as > possible -- i.e. extend the previous frame by "length" bytes until you see > END_HEADERS flag. I could be convinced to accept that reasoning - fewer places to look for bits seems like a good thing. I was hoping that it would be the last frame, but given the nature of continuations, this is tempting. What do others think? END_HEADERS is on HEADERS + PUSH_PROMISE + CONTINUATION, but END_STREAM is only on HEADERS. Later, END_MESSAGE might only be on HEADERS or DATA, depending on how things work out.
Received on Monday, 12 August 2013 23:59:05 UTC