[Bug 30296] Add Addressing support to Gabble

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=30296

--- Comment #2 from Eitan Isaacson <eitan.isaacson at collabora.co.uk> 2010-09-22 21:05:48 PDT ---
Ok, I put into this almost a full week. And it is still far from perfect.
This really needs to end up in tp-glib, any work done in gabble (which is not a
trivial delta) will pretty much be thrown away once it goes into tp-glib.

I'm not going to bother rebasing this, unless you think this is something we
could actually introduce into master.

Some random thoughts:
* The protocol stuff could go into TpBaseProtocol.
* Static gchar* arrays for interfaces make it extremely unflexible. It worked
fine when you just do +1 for removing a group iface, but what about two
mutually exclusive interfaces? like group and addressing. There is probably a
better way of dealing with this, but my current branch just has random
undocumented +1s.
* I hope I understood RCCs. From what I gather, every RCC that had a fixed
handle type of contact there will now be two more, one for vcard addresses and
one for URIs. This is an extremely large amount of RCCs! Most of the caps
related tests are failing now, they need to be more streamlined to support this
multiplication. Same goes with all the foreach_channel_class functions, they
need to multiply all the 1-on-1 classes. I set up a helper function for that. I
think that tp-glib should probably do that automatically when it builds the
RCCs (if the cm supports addressing).
* We never talked about how addressing works with MUCs, for now it just
doesn't. (ie, a TargetVcardAddress request will assume a TargetHandleType of
contact).
* tp-glib does not allow having a handle type without a handle or identifier in
a channel request. So for now that means that using the addressing bits for
requests requires handle type to be omitted.
* Some 20 odd tests are still failing, they are mostly caps related, and can't
deal with the RCCs being so different.

That's it, letting go for now...

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.