- From: 陈智昌 <willchan@chromium.org>
- Date: 2013年8月14日 00:05:22 +0200
- To: James M Snell <jasnell@gmail.com>
- Cc: ietf-http-wg@w3.org, Tatsuhiro Tsujikawa <tatsuhiro.t@gmail.com>, Martin Thomson <martin.thomson@gmail.com>
- Message-ID: <CAA4WUYhdE77vxw_rDtcgWf1R5U2x7cT28MPkkcmFjjvz-+KasQ@mail.gmail.com>
http://www.chromium.org/spdy/spdy-proxy The CONNECT tunnel is implemented as Tatsuhiro-san described (the mapping is admittedly unclean). There are some complexities when you tunnel SPDY over SPDY due to multiplexing and flow control, but otherwise I think the rest is relatively straightforward, and indeed there are open source SPDY proxy implementations that support this like Tatsuhiro-san's. On Aug 13, 2013 11:51 PM, "James M Snell" <jasnell@gmail.com> wrote: > I mean that, at some point, we ought to sit down and figure out a > better approach to handling this case because using CONNECT as > currently defined is really not a great fit and I'd rather not try to > just shoe horn something in with a bunch of new exception cases. If > Google has this working, please provide a detailed description on list > of what's being done so we can evaluate it. >
Received on Tuesday, 13 August 2013 22:12:37 UTC