Does the the web polygraph use http/1.1?

I'm seeing pretty good client-side stats now (30+ requests per connection) (I've used the cachemgr 68 times :-]) The <10 requests
where with http/1.0 persistent connections before I told IE to use http/1.1

I'm wondering how the te code would cope with a polygraph test? If anyone has the time and inclination I'd love to hear how it goes.
The code is ugly and certainly not release ready, but from here on in it's fine tuning only..

It currently always decodes the server side so we store an identity entity. Unfortunately (as per my email "storemanager") the
stored entity still has it's connection and transfer-encoding headers (this is a bug for all current squids AFAICT).

But the client-side code is able to skip doing a memcopy IFF no client side transfer encodings are present (ie for http/1.0
clients).

Rob

=
Client-side persistent connection counts:

        req/
        conn count
        ---- ---------
           1 68
           2 4
           3 3
           4 3
           5 1
           6 3
           7 1
           9 1
          10 1
          11 2
          13 1
          14 1
          15 1
          18 2
          19 1
          25 1
          29 1
          31 1
          32 1
Received on Sat Jan 13 2001 - 02:27:39 MST