4 - Audit @todo and FIXME for security issues
6 - The convenience functions in dbus-bus.h should perhaps have
7 the signatures that they would have if they were autogenerated
8 stubs. e.g. the acquire service function. We should also evaluate
9 which of these functions to include, in light of the fact that
10 GLib/Qt native stubs will probably also exist.
12 - the "break loader" and valid/invalid message tests are all disabled;
13 they need to be fixed and re-enabled with the new message args stuff.
14 I think I want to drop the .message files thing and just have code
15 that generates messages, more like the tests for
16 dbus-marshal-recursive.c (this is mostly done now, just needs some
19 - need to define bus behavior if you send a message to
20 yourself; is it an error, or allowed? If allowed,
21 we need to have a test for it in the test suite.
23 - validate dict entry number of fields
25 - just before 1.0, try a HAVE_INT64=0 build and be sure it runs
27 - in dbus-keyring.c, enforce that the keyring dir is not
28 world readable/writable
32 - dbus-pending-call.c has some API and thread safety issues to review
34 - Add test harness for selinux allow/deny cf. this message
35 http://lists.freedesktop.org/archives/dbus/2005-April/002506.html
37 Important for 1.0 GLib Bindings
40 - finish dbus-glib-tool support for adding introspection
41 data to GObject and autoexporting GObject using same
43 - Need to make sure that dbus-glib.h never returns any
44 dbus_malloc() memory, only g_malloc().
45 dbus_g_proxy_end_call() is the major offender.
47 - DBusGProxy doesn't emit "destroy" when it should
52 - add dbus_message_has_path(), maybe has_member/interface
54 - connection_open/connection_disconnect lacks symmetry, open/close
57 - protocol version in each message is pretty silly
62 - if the GUID is obtained only during authentication, not in the address,
63 we could still share the connection
65 - Allow a dbus_g_proxy_to_string()/g_object_to_string() that
66 would convert the proxy to an "IOR" and dbus_g_proxy_from_string()
67 that would decode; using these, dbus-glib users could avoid
68 DBusConnection entirely. Of course the same applies to other kinds
69 of binding. This would use dbus_connection_open()'s connection-sharing
70 feature to avoid massive proliferation of connections.
72 - DBusWatchList/TimeoutList duplicate a lot of code, as do
73 protected_change_watch/protected_change_timeout in dbus-connection.c
74 and dbus-server.c. This could all be mopped up, cut-and-paste
75 fixed, code size reduced.
77 - change .service files to allow Names=list in addition to Name=string
79 - The message bus internal code still says "service" for
80 "name", "base service" for "unique name", "activate" for
81 "start"; would be nice to clean up.
83 - Property list feature on message bus (list of properties associated
84 with a connection). May also include message matching rules
85 that involve the properties of the source or destination
88 - Disconnecting the remote end on invalid UTF-8 is probably not a good
89 idea. The definition of "valid" is slightly fuzzy. I think it might
90 be better to just silently "fix" the UTF-8, or perhaps return an error.
92 - build and install the Doxygen manual in Makefile when --enable-docs
94 - if you send the same message to multiple connections, the serial number
95 will only be right for one of them. Probably need to just write() the serial
96 number, rather than putting it in the DBusMessage, or something.
98 - perhaps the bus driver should have properties that reflect attributes
99 of the session, such as hostname, architecture, operating system,
100 etc. Could be useful for code that wants to special-case behavior
101 for a particular host or class of hosts, for example.
103 - currently the security policy stuff for messages to/from
104 the bus driver is kind of strange; basically it's hardcoded that
105 you can always talk to the driver, but the default config file
106 has rules for it anyway, or something. it's conceptually
107 screwy at the moment.
109 - when making a method call, if the call serial were globally unique,
110 we could forward the call serial along with any method calls made
111 as a result of the first method call, and allow reentrancy that was
112 strictly part of the call stack of said method call. But I don't
113 really see how to do this without making the user pass around the
114 call serial to all method calls all the time, or disallowing
117 If done post 1.0 will probably be an optional/ugly-API type
120 - I don't want to introduce DBusObject, but refcounting and object
121 data could still be factored out into an internal "base class"
124 - document the auth protocol as a set of states and transitions, and
125 then reimplement it in those terms
127 - recursive dispatch, see dbus_connection_dispatch()
129 - do we need per-display activation; if so I'd like to do this by setting a
130 "display ID" property on screen 0, with a GUID, and keying activation by
131 said GUID. Otherwise you get all kinds of unrobust
132 string/hostname-based mess. per-screen is then done by appending screen number
133 to the display. If displays have a deterministic ID like this, you can
134 do per-display by simply including GUID in the service name.
136 - optimization and profiling!