New regression test: dynamic forwarding

[Markus Friedl]

On Thu, Jun 26, 2003 at 03:40:57AM -0700, Dan Kaminsky wrote:
> Given that we already have a good portion of the necessary code to 
> support simple SOCKS already, _and_ that a built-in client would then be 
> able to interface with our already built-in server, why _not_ support an 
> ultra-simple socks4/socks5 client mode?
> 
> I mean, we already have the struct declaration, and it's only a few 

but that's it.

perhaps.