On Mon, Nov 14, 2016 at 07:44:26PM -0500, Jeff King wrote: > > But it does seem to work. At least it doesn't seem to break anything in > > the test suite, and it fixes the new tests you added. I'd worry that > > there's some obscure case where the response isn't packetized in the > > same way. > > Actually, I take it back. I think it works for a single round of ref > negotiation, but not for multiple. Enabling GIT_TEST_LONG=1 causes it to > fail t5551. > > I think I've probably made a mis-assumption on exactly when in the HTTP > protocol we will see a flush packet (and perhaps that is a sign that > this protocol-snooping approach is not a good one). Yep, that is it. The server may end with an ACK or a NAK, not a flush packet. So going this route really does mean teaching remote-curl a lot more about the protocol, which is pretty nasty. I think trying to somehow signal the end-of-response to fetch-pack would be a more maintainable approach. -Peff