6 - the README, FAQ, and probably other places talk about 1.0 and ABI guarantees,
7 all these places need updating
9 - the spec should say "the protocol will not be changed incompatibly after Month DD, YYYY"
11 Important for 1.0 GLib Bindings
14 - Test point-to-point mode
16 - Add support for getting sender
18 - format_version in the object info doesn't look like it's handled correctly. The creator
19 of the object info should specify some fixed number per struct version; the library
20 should handle only specific numbers it knows about. There's no assumption that all
21 numbers >= the given one are compatible. The idea is that new versions of the lib
22 can offer totally different object info structs, but old versions
25 Important for 1.0 Python bindings
30 - Fix removing of signals from the match tree
32 - Fix refcounting and userdata lifecycles
34 - Write a generic mainloop
39 - protocol version in each message is pretty silly
44 - revamp dbus-launch a bit,
45 see http://lists.freedesktop.org/archives/dbus/2006-October/005906.html
48 - clean up the creds issue on *BSD's in dbus/dbus-sysdeps-unix.c.
49 They should work as is but we need to rearange it to make it
50 clearer which method is being used. configure.in should
51 be fixed up to make that decition.
53 - _dbus_connection_unref_unlocked() is essentially always broken because
54 the connection finalizer calls non-unlocked functions. One fix is to make
55 the finalizer run with the lock held, but since it calls out to the app that may
56 be pretty broken. More likely all the uses of unref_unlocked are just wrong.
58 - if the GUID is obtained only during authentication, not in the address,
59 we could still share the connection
61 - Allow a dbus_g_proxy_to_string()/g_object_to_string() that
62 would convert the proxy to an "IOR" and dbus_g_proxy_from_string()
63 that would decode; using these, dbus-glib users could avoid
64 DBusConnection entirely. Of course the same applies to other kinds
65 of binding. This would use dbus_connection_open()'s connection-sharing
66 feature to avoid massive proliferation of connections.
68 - DBusWatchList/TimeoutList duplicate a lot of code, as do
69 protected_change_watch/protected_change_timeout in dbus-connection.c
70 and dbus-server.c. This could all be mopped up, cut-and-paste
71 fixed, code size reduced.
73 - change .service files to allow Names=list in addition to Name=string
75 - The message bus internal code still says "service" for
76 "name", "base service" for "unique name", "activate" for
77 "start"; would be nice to clean up.
79 - Property list feature on message bus (list of properties associated
80 with a connection). May also include message matching rules
81 that involve the properties of the source or destination
84 - Disconnecting the remote end on invalid UTF-8 is probably not a good
85 idea. The definition of "valid" is slightly fuzzy. I think it might
86 be better to just silently "fix" the UTF-8, or perhaps return an error.
88 - build and install the Doxygen manual in Makefile when --enable-docs
90 - if you send the same message to multiple connections, the serial number
91 will only be right for one of them. Probably need to just write() the serial
92 number, rather than putting it in the DBusMessage, or something.
94 - perhaps the bus driver should have properties that reflect attributes
95 of the session, such as hostname, architecture, operating system,
96 etc. Could be useful for code that wants to special-case behavior
97 for a particular host or class of hosts, for example.
99 - currently the security policy stuff for messages to/from
100 the bus driver is kind of strange; basically it's hardcoded that
101 you can always talk to the driver, but the default config file
102 has rules for it anyway, or something. it's conceptually
103 screwy at the moment.
105 - when making a method call, if the call serial were globally unique,
106 we could forward the call serial along with any method calls made
107 as a result of the first method call, and allow reentrancy that was
108 strictly part of the call stack of said method call. But I don't
109 really see how to do this without making the user pass around the
110 call serial to all method calls all the time, or disallowing
113 If done post 1.0 will probably be an optional/ugly-API type
116 - I don't want to introduce DBusObject, but refcounting and object
117 data could still be factored out into an internal "base class"
120 - Keep convenience wrappers in sync with bus methods
122 - document the auth protocol as a set of states and transitions, and
123 then reimplement it in those terms
125 - recursive dispatch, see dbus_connection_dispatch()
127 - do we need per-display activation; if so I'd like to do this by setting a
128 "display ID" property on screen 0, with a GUID, and keying activation by
129 said GUID. Otherwise you get all kinds of unrobust
130 string/hostname-based mess. per-screen is then done by appending screen number
131 to the display. If displays have a deterministic ID like this, you can
132 do per-display by simply including GUID in the service name.
134 - optimization and profiling!
136 - Match rules aren't in the spec (probably a lot of methods on the bus
139 - the "break loader" and valid/invalid message tests are all disabled;
140 they need to be fixed and re-enabled with the new message args stuff.
141 I think I want to drop the .message files thing and just have code
142 that generates messages, more like the tests for
143 dbus-marshal-recursive.c (this is mostly done now, just needs some
146 - just before 1.0, try a HAVE_INT64=0 build and be sure it runs
148 - Windows port needs recursive mutexes
153 - look into supporting the concept of a "connection" generically
154 (what does this TODO item mean?)
156 - test/name-test should be named test/with-bus or something like that