I'm trying to figure out what changes are needed on the Java side.
It seems the bind method will be of interest.

Rafi, could you also explain how the refactor is going to help with SSL ?

Rajith

On Tue, Sep 4, 2012 at 5:09 PM, Rafael Schloming <r...@alum.mit.edu> wrote:

> Hi Everyone,
>
> I've done some exploratory coding on a small refactor aimed at simplifying
> the driver in order to ease both portability and ssl integration. I've
> attached the patch. It's not huge, but it removes some of the trickiest
> code from driver.c. I believe there are further simplifications that could
> be made on top of this, but I think this represents the minimal coherent
> step. This involves the following API modifications:
>
> Changed:
>   pn_sasl_t *pn_sasl() -> pn_sasl_t *pn_sasl(pn_transport_t *transport)
>   pn_transport_t *pn_transport(pn_connection_t *connection) ->
> pn_transport_t *pn_transport(void)
>
> Added:
>   int pn_transport_bind(pn_transport_t *transport, pn_connection_t
> *connection)
>
> Note that a transport can now exist independently from a connection and
> must therefore also be explicitly freed.
> Note that the sasl object does *not* need to be explicitly freed as it
> will be automatically freed when the transport is freed.
>
> If you're using the messenger API or the driver and engine API together,
> then there will be no visible changes.
>
> Please have a look and give me any comments I'll be tying up some loose
> ends and looking to commit this soon.
>
>
> --Rafael
>

Reply via email to