Alex Rousskov wrote:
>
> I do not know what error you are getting, but it is possible that
> auto-conversion from HttpRequestMethod to method_t is at fault. If so,
> we should probably try to remove that conversion.
>
> Again, let me know if you need a hand.
>
> Thank you,
>
Alex, I'm droppign the conversion and not seeing many uses of METHOD_*
as an int. the ones present however look to be important a keepers.
You mentioned earlier when this class went it it might be an idea to
look at removing the _method_t enum entirely.
After today I don't think it would be a good idea for performance. BUT,
removing many of the entries to it which are unused would be a good
thing IMO. That would speed up the method/URI parsing very slightly on
(rare?) occasions and reduce the code a fair bit.
I'm thinking keeping GET, HEAD, POST, etc and any which are need for
fast path switching in the code.
What do you think of that as a next-step polish?
Amos
-- Please use Squid 2.6STABLE17+ or 3.0STABLE1+ There are serious security advisories out on all earlier releases.Received on Sat Feb 02 2008 - 16:11:34 MST
This archive was generated by hypermail pre-2.1.9 : Sat Mar 01 2008 - 12:00:09 MST