Chuck Murcko wrote: > That's me remembrance. The new mod_proxy client still doesn't do > (de)chunking, for instance. The 2.0 mod_proxy does. Perhaps we were more > correct with the client version as it was?
Er... in my original patch the version level was HTTP/1.1 - I have no
idea how it got changed to v1.0 again.
As far as I am aware, the buff.c code handles dechunking of a request or
response - can someone confirm this for me?
> The original thinking was that further effort beyond this upgrade (the
> 1.3.22 mod_proxy change) and its maintenance would be better spent in
> 2.0. Is it time to rethink that?
The upgrades to both v1.3 and 2.0 are complete bar bugfixes and one
protocol issue (Cache-control: no-cache=<header>). There is no point in
not fixing either v1.3 or v2.0 if problems exist. In addition, noone has
yet posted a specific instance where the v1.3 proxy violates RFC2616. If
someone does find an instance where it violates the spec, mail me and I
(or someone else) will fix it.
In the meantime I don't think downgrading the proxy is justified without
a concrete reason to do so.
Regards,
Graham
--
-----------------------------------------
[EMAIL PROTECTED] "There's a moon
over Bourbon Street
tonight..."
smime.p7s
Description: S/MIME Cryptographic Signature
