Yes, looks like the Commons HTTP (or some other HTTP library) could provide a solution. The nice thing about replacing socket implementations was that it was a general solution applicable also to other TCP-based protocols (e.g. JDBC).

--
Salut,

Jordi.

Martin Ramshaw wrote:
(I re-signed up for the developer list, please reply to the list).


I said I would research the possiblity of changing the socket factory by
a custom one that would produce sockets capable of measuring times below
the HTTP level (connection establishment, write time, latency, read time).


Some time ago, at Mike Stovers' suggestion, I looked at the possibility of
using the Commons HTTP classes. A quick trial convinced me that the work
involved was massive and well beyond the time that I could afford to
contribute.

But if they will do what you require, I'd be prepared to help out.

<snip>

Most of these objections were addressed by the Commons people. In
addition, they really serve to justify the Commons HTTP project. There is
also a Swiss effort that I looked at due to a requirement to be able to force
URLConnections to time-out (now, I believe, addressed by Java.NIO).


In short: java.net was not designed to be extensible by common mortals,


Agreed. The original implementation was a bit of a hack; there are lots of
bugs reported against it which Sun has been very reluctant to address.


and a Real Java Programmer(tm) is needed for the job :-)


Don't be so modest!


If you can come up with an idea to solve this (other than repackaging
rt.jar -- that is too dirty even for me), I'll be happy to explore it.


See above.

Regards


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



-- Salut,

Jordi.


--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to