4 - add a new return code from dbus_connection_dispatch() called
5 IN_PROGRESS or RECURSED or something, indicating that DATA_REMAINS
6 but another dispatch is in progress, so we can't dispatch at
7 this time. OR maybe just switch to recursive locks for the dispatch
8 locks. Fixes the recursive deadlock. See the @todo for more
9 and this thread: http://lists.freedesktop.org/archives/dbus/2006-February/004128.html
11 - Remove all deprecated functions
13 - Audit @todo and FIXME for security issues
15 - the "break loader" and valid/invalid message tests are all disabled;
16 they need to be fixed and re-enabled with the new message args stuff.
17 I think I want to drop the .message files thing and just have code
18 that generates messages, more like the tests for
19 dbus-marshal-recursive.c (this is mostly done now, just needs some
22 - just before 1.0, try a HAVE_INT64=0 build and be sure it runs
24 - dbus-pending-call.c has some API and thread safety issues to review.
25 DBusPendingCall is used from multiple threads with no locks.
26 Either DBusConnection's lock has to protect all associated pending
27 call (means pending->connection can't ever be set to null) or
28 or DBusPendingCall needs its own lock
29 http://lists.freedesktop.org/archives/dbus/2006-June/004945.html
31 - Add test harness for selinux allow/deny cf. this message
32 http://lists.freedesktop.org/archives/dbus/2005-April/002506.html
34 - publish the introspection dtd at its URL
36 - RequestName flags seem a bit strange; see the docs for dbus_bus_request_name()
37 and think about use cases in better detail.
39 http://lists.freedesktop.org/archives/dbus/2005-August/003207.html
41 Kind of a major API change, but seems high-value.
43 - dbus_bus_get() should hold a strong reference associated with the "connected"
44 state (i.e. libdbus drops its reference when the connection disconnects,
45 and sets its internal connection variable to null).
46 See http://lists.freedesktop.org/archives/dbus/2006-May/004806.html
48 Important for 1.0 GLib Bindings
51 - Test point-to-point mode
53 - Add support for getting sender
55 - format_version in the object info doesn't look like it's handled correctly. The creator
56 of the object info should specify some fixed number per struct version; the library
57 should handle only specific numbers it knows about. There's no assumption that all
58 numbers >= the given one are compatible. The idea is that new versions of the lib
59 can offer totally different object info structs, but old versions
62 Important for 1.0 Python bindings
67 - Fix removing of signals from the match tree
69 - Fix refcounting and userdata lifecycles
71 - Write a generic mainloop
76 - protocol version in each message is pretty silly
81 - if the GUID is obtained only during authentication, not in the address,
82 we could still share the connection
84 - Allow a dbus_g_proxy_to_string()/g_object_to_string() that
85 would convert the proxy to an "IOR" and dbus_g_proxy_from_string()
86 that would decode; using these, dbus-glib users could avoid
87 DBusConnection entirely. Of course the same applies to other kinds
88 of binding. This would use dbus_connection_open()'s connection-sharing
89 feature to avoid massive proliferation of connections.
91 - DBusWatchList/TimeoutList duplicate a lot of code, as do
92 protected_change_watch/protected_change_timeout in dbus-connection.c
93 and dbus-server.c. This could all be mopped up, cut-and-paste
94 fixed, code size reduced.
96 - change .service files to allow Names=list in addition to Name=string
98 - The message bus internal code still says "service" for
99 "name", "base service" for "unique name", "activate" for
100 "start"; would be nice to clean up.
102 - Property list feature on message bus (list of properties associated
103 with a connection). May also include message matching rules
104 that involve the properties of the source or destination
107 - Disconnecting the remote end on invalid UTF-8 is probably not a good
108 idea. The definition of "valid" is slightly fuzzy. I think it might
109 be better to just silently "fix" the UTF-8, or perhaps return an error.
111 - build and install the Doxygen manual in Makefile when --enable-docs
113 - if you send the same message to multiple connections, the serial number
114 will only be right for one of them. Probably need to just write() the serial
115 number, rather than putting it in the DBusMessage, or something.
117 - perhaps the bus driver should have properties that reflect attributes
118 of the session, such as hostname, architecture, operating system,
119 etc. Could be useful for code that wants to special-case behavior
120 for a particular host or class of hosts, for example.
122 - currently the security policy stuff for messages to/from
123 the bus driver is kind of strange; basically it's hardcoded that
124 you can always talk to the driver, but the default config file
125 has rules for it anyway, or something. it's conceptually
126 screwy at the moment.
128 - when making a method call, if the call serial were globally unique,
129 we could forward the call serial along with any method calls made
130 as a result of the first method call, and allow reentrancy that was
131 strictly part of the call stack of said method call. But I don't
132 really see how to do this without making the user pass around the
133 call serial to all method calls all the time, or disallowing
136 If done post 1.0 will probably be an optional/ugly-API type
139 - I don't want to introduce DBusObject, but refcounting and object
140 data could still be factored out into an internal "base class"
143 - Keep convenience wrappers in sync with bus methods
145 - document the auth protocol as a set of states and transitions, and
146 then reimplement it in those terms
148 - recursive dispatch, see dbus_connection_dispatch()
150 - do we need per-display activation; if so I'd like to do this by setting a
151 "display ID" property on screen 0, with a GUID, and keying activation by
152 said GUID. Otherwise you get all kinds of unrobust
153 string/hostname-based mess. per-screen is then done by appending screen number
154 to the display. If displays have a deterministic ID like this, you can
155 do per-display by simply including GUID in the service name.
157 - optimization and profiling!
159 - Match rules aren't in the spec (probably a lot of methods on the bus
165 - look into supporting the concept of a "connection" generically