Expose Multiple Resources

Jason Salaz jason at zenenet.com
Tue Jul 24 19:17:54 UTC 2007

On 7/24/07, Martin Atkins <mart at degeneration.co.uk> wrote:
> Peter Saint-Andre wrote:
> > Jason Salaz wrote:
> >> On 7/24/07, Martin Atkins <mart at degeneration.co.uk> wrote:
> >>> Jason Salaz wrote:
> >>>  > I would imagine that the problem is that the server doesn't allow
> >>>  > presence to<->from oneself, or something like that that.
> >>>
> >>> Do you know how this is supposed to work at the protocol level? If
> >>> you're not sure then I'll try to figure it out myself, of course. :)
> >> I'm not entirely sure.  I'm looking through rfc3920, but that's
> >> intermingled with tending to my day job as well, so I may not get to
> >> the bulk of it until tonight.
> >>
> >> What I do know is that I don't have any subscription to myself. i.e.
> >> My JID isn't on my own roster as a subscription.  Yet, GTalk still
> >> passes my own presence from another resource along whenever I sign on.
> >
> > This is a feature of the protocol. It is best explained here:
> >
> > http://www.xmpp.org/internet-drafts/draft-saintandre-rfc3921bis-03.html#presence-initial-outbound
> >
> > Please refer to rfc3920bis and rfc3921bis for the most up-to-date
> > descriptions of the protocol.
> >
> I have coded up a fix. The change is that whenever the server does a
> presence broadcast to everyone on the roster it also sends the same
> presence stanza to all of the user's other resources.
> This seems to satisfy what the spec says, and makes Psi show the "Send
> Message To" menu item for the other resources.
> However, this has caused DJabberd to no longer pass the "two-resources"
> tests. I'll be checking the fix in once I've figured out why the tests
> are now failing.

What are the details of this "two-resources" test? What is expected,
what is the result?

More information about the Djabberd mailing list