[Bug 29077] refactor the roster, moving logic out of the channels into the roster itself

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=29077

--- Comment #2 from Simon McVittie <simon.mcvittie at collabora.co.uk> 2010-09-22 08:16:10 PDT ---
(In reply to comment #1)
> Test deleting contact lists via Close()
> 
> Is this change merged back to the corresponding test in Gabble?

There's no directly corresponding test, so, no; libpurple's contact list
handling differs sufficiently that the tests would have to differ anyway. My
TpBaseContactList branch for Gabble does add a test for deleting non-empty
groups.

> +    bus.get_object(conn.bus_name, '/').Ping()
> 
> sync_dbus?

Obvious change applied.

> haze_contact_list_add_to_group() uses:
> 
> +    PurpleGroup *group = purple_group_new (group_name);
> 
> which is actually kosher, because _new() returns an existing group if possible.
> It threw me a bit, though. Guess I've been away too long. Any chance of a
> comment explaining that this is correct?

    /* This is correct, despite the naming: it returns a borrowed reference
     * to an existing group (if possible) or to a new group (otherwise). */
    PurpleGroup *group = purple_group_new (group_name);

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