ws-soap-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From George I Matkovits <>
Subject Re: HTTP protocol implementation
Date Sat, 15 Jul 2000 01:21:22 GMT
I should have said: I64 encoded and serialized Java Objects below ....
Sorry - George

George I Matkovits wrote:

> Please also look at the Java2 APIs for HTTP support from inside servlets. Most
> good servlet books cover http tunneling and things like cookies, session
> management, authentication, PKI managed SSL support with JSSE is not as widely
> documented yet but can be made to work.
> It is NOT just XML all the way!
> Serialized Java objects can be directly transmitted over http without platform
> dependent prints. In addition, what Soap will provide is the option for a
> consistent standard of  an end to and interface. The XML based interface
> definition will allow (eventually) true B2B, when the other requirements like
> data privacy and security will be also satisfied. If we here do not find a
> 'generic' and open solution for the security issues, SOAP will not make it IMHO
> and the Internet standard's body IETF will be open to billion dollar law suits.
> Regards -George
> Michael Brennan wrote:
> > Here's another implementation question I have. (Again, I hope folks don't
> > mind my asking.)
> >
> > The current HTTP protocol support seems very lightweight. What are the plans
> > going forward? Any consideration of using third party HTTP lib? I've been
> > looking, for instance, at the HTTP client library from the W3C included in
> > the Jigsaw distribution. Rich support for HTTP/1.1 including authentication,
> > cookies, chunked encoding, etc. Also has a very extensible architecture. For
> > instance, the ability to attach filters that can modify requests or
> > responses (this is how support for cookies, authentication, etc, is
> > implemented). It seems like there is some reinventing of the wheel going on
> > with the protocol support.
> >
> > I also noticed that the method creates a PrintWriter for
> > output on top of the output stream. That results in the local system default
> > character encoding being used. Isn't this a bit dangerous? Shouldn't this be
> > explicitly specifying, perhaps, a utf-8 encoding? Likewise with the Reader
> > for reading the response.
> >
> > I'm very much a novice with SOAP and this implementation, so I apologize in
> > advance if these are naive questions.
> >
> > Thanks.

View raw message