On Thu, 5 Sep 2013, venkatesh perumalla wrote:
Little explanation in below link. http://stackoverflow.com/questions/10480122/difference-between-http-pipeling-and-http-multiplexing-with-spdy
Thanks but I don't need neither spdy nor http2 education. I've been on both mailing lists and kept up with their developments since they started.
In case of pipeline mode, it is gaurantees that we get in the same sequence.
I don't see why that is relevant to us. Sure it makes that if you start request A first then B, request B may complete before A. That's not really a problem in the libcurl API since we have never claimed to do them in any order. Including when "plain" HTTP pipeline is enabled and used.
In case of http2 or spdy, if we use same multi interface, if we read from socket, it would have come for E2 or E3.
Exactly as if you add two easy handles using two different connections. To a user it won't be a difference!
[snipped in from the second email]
I see there are somany challenges. If we use same interface.
I'm sorry, but I see even more challenges if we were to introduce yet another interface and one that would be added only(?) for http2 support...
-- / daniel.haxx.se ------------------------------------------------------------------- List admin: http://cool.haxx.se/list/listinfo/curl-library Etiquette: http://curl.haxx.se/mail/etiquette.html
