2 - Message matching rules (so broadcasts can be filtered) need sorting
5 - How we will handle DCOP needs sorting out. Among other things, we
6 need to check that service and service-ownership semantics map to DCOP
9 - Activation needs some careful additional thinking-through.
11 - Property list feature on message bus (list of properties associated
12 with a connection). May also include message matching rules
13 that involve the properties of the source or destination
16 - Automatic service activation, should probably be done through a message flag.
18 - Disconnecting the remote end on invalid UTF-8 is probably not a good
19 idea. The definitiion of "valid" is slightly fuzzy. I think it might
20 be better to just silently "fix" the UTF-8, or perhaps return an error.
22 Owen says we should only validate the UTF-8 on dbus_message_get_string()
23 (changing get_string to have an error return, and allowing a type error
26 - The convenience functions in dbus-bus.h should perhaps have
27 the signatures that they would have if they were autogenerated
28 stubs. e.g. the acquire service function. We should also evaluate
29 which of these functions to include, in light of the fact that
30 GLib/Qt native stubs will probably also exist.
32 - The message handler interface needs rethinking, perhaps handlers should be able
33 to return an error that automatically gets turned into a message; most likely
34 some basic spec'ing out of the GLib/Qt level stubs/skels stuff will be
35 needed to understand the right approach.
37 - there are various bits of code to manage ref/unref of data slots, that should
38 be merged into a generic facility
40 - assorted _-prefixed symbols in libdbus aren't actually used by
41 libdbus, only by the message bus. These bloat up the library
42 size. Not sure how to fix, really.
44 - if you send a message to a service then block for reply, and the service exits/crashes
45 after the message bus has processed your message but before the service has replied,
46 it would be nice if the message bus sent you an error reply.
48 - build and install the Doxygen manual in Makefile when --enable-docs
50 - if you send the same message to multiple connections, the serial number
51 will only be right for one of them. Probably need to just write() the serial
52 number, rather than putting it in the DBusMessage, or something.
54 - perhaps the bus driver should have properties that reflect attributes
55 of the session, such as hostname, architecture, operating system,
56 etc. Could be useful for code that wants to special-case behavior
57 for a particular host or class of hosts, for example.
59 - currently the security policy stuff for messages to/from
60 the bus driver is kind of strange; basically it's hardcoded that
61 you can always talk to the driver, but the default config file
62 has rules for it anyway, or something. it's conceptually
65 - <limit> elements are not merged in from included configuration
66 files; they have to be in the toplevel file. when loading
67 a child file, we could just init its DBusLimits from the parent,
68 then after parsing copy its DBusLimits back to the parent
70 - when making a method call, if the call serial were globally unique,
71 we could forward the call serial along with any method calls made
72 as a result of the first method call, and allow reentrancy that was
73 strictly part of the call stack of said method call. But I don't
74 really see how to do this without making the user pass around the
75 call serial to all method calls all the time, or disallowing
78 - the invalid messages in the test suite are all useless because
79 they are invalid for the wrong reasons due to protocol changes
81 - I don't want to introduce DBusObject, but refcounting and object
82 data could still be factored out into an internal "base class"
85 - modify the auth protocol to also support other initial-handshake
88 - document the auth protocol as a set of states and transitions, and
89 then reimplement it in those terms
91 - dbus_gproxy or dbus_g_proxy?
93 - add dbus_message_has_path(), maybe has_member/interface
95 - The OBJECT_PATH type is not documented in the spec.