Better HTTP 1.1 support in http.server?
Hello, I was wondering if it would be worth providing better HTTP 1.1 support in http.server. The way I envision it, there would be a separate HTTP11RequestHandler which would provide: - a smart wfile with automatic chunk encoding (which relieves the API user from manually handling chunk encoding or content length calculation) - keep-alive enabled by default - HTTP 1.1 by default (BaseHTTPRequestHandler defaults to 1.0) What do you think? Regards Antoine. -- Software development and contracting: http://pro.pitrou.net
Am 24.09.2012 17:27, schrieb Antoine Pitrou:
Hello,
I was wondering if it would be worth providing better HTTP 1.1 support in http.server. The way I envision it, there would be a separate HTTP11RequestHandler which would provide:
- a smart wfile with automatic chunk encoding (which relieves the API user from manually handling chunk encoding or content length calculation) - keep-alive enabled by default - HTTP 1.1 by default (BaseHTTPRequestHandler defaults to 1.0)
What do you think?
+1 for 3.4 You proposed gave me another idea. What do you think about SPDY support in the stdlib? It's the next step after HTTP 1.1.
On Mon, Sep 24, 2012 at 6:39 PM, Christian Heimes <christian@python.org> wrote:
You proposed gave me another idea. What do you think about SPDY support in the stdlib? It's the next step after HTTP 1.1.
I'd wait it out a bit. SPDY is currently iterating, and there's an effort to define HTTP 2 that will likely supersede SPDY (and may incorporate many of its ideas). http://trac.tools.ietf.org/wg/httpbis/trac/wiki/Http2CfI On the other hand, some WebSockets support might be useful. Cheers, Dirkjan
On Mon, Sep 24, 2012 at 9:39 AM, Christian Heimes <christian@python.org> wrote:
+1 for 3.4
You proposed gave me another idea. What do you think about SPDY support in the stdlib? It's the next step after HTTP 1.1.
Yeah, it is a good idea. it should live along side with HTTP 1.1 as another protocol layer. Guido had given some suggestions to takle that, but I have been lazing and lagging for while now in terms of contributions, but I am excited about the idea. -- Senthil
+1 for 3.4 And I'll be available to help with HTTP(websockets and SPDY) related stuff! 2012/9/24 Senthil Kumaran <senthil@uthcode.com>
On Mon, Sep 24, 2012 at 9:39 AM, Christian Heimes <christian@python.org> wrote:
+1 for 3.4
You proposed gave me another idea. What do you think about SPDY support in the stdlib? It's the next step after HTTP 1.1.
Yeah, it is a good idea. it should live along side with HTTP 1.1 as another protocol layer. Guido had given some suggestions to takle that, but I have been lazing and lagging for while now in terms of contributions, but I am excited about the idea.
-- Senthil _______________________________________________ Python-Dev mailing list Python-Dev@python.org http://mail.python.org/mailman/listinfo/python-dev Unsubscribe: http://mail.python.org/mailman/options/python-dev/felipecruz%40loogica.net
On 24 September 2012 19:39, Christian Heimes <christian@python.org> wrote:
Am 24.09.2012 17:27, schrieb Antoine Pitrou:
Hello,
I was wondering if it would be worth providing better HTTP 1.1 support in http.server. The way I envision it, there would be a separate HTTP11RequestHandler which would provide:
- a smart wfile with automatic chunk encoding (which relieves the API user from manually handling chunk encoding or content length calculation) - keep-alive enabled by default - HTTP 1.1 by default (BaseHTTPRequestHandler defaults to 1.0)
What do you think?
+1 for 3.4
You proposed gave me another idea. What do you think about SPDY support in the stdlib? It's the next step after HTTP 1.1.
Given the rate SPDY is changing, complexity of implementation, and that some prerequisites aren't in place (zlib module needs to support specifying the dictionary), this probably isn't a good idea right away. It's also easy to get a similar effect using a SPDY-capable reverse proxy.
_______________________________________________ Python-Dev mailing list Python-Dev@python.org http://mail.python.org/mailman/listinfo/python-dev Unsubscribe: http://mail.python.org/mailman/options/python-dev/dw%40botanicus.net
participants (6)
-
Antoine Pitrou
-
Christian Heimes
-
David Wilson
-
Dirkjan Ochtman
-
Felipe Cruz
-
Senthil Kumaran