Find a file
Havoc Pennington dbdea921b5 2004-11-25 Havoc Pennington <hp@redhat.com>
The primary change here is to always write() once before adding
	the write watch, which gives us about a 10% performance increase.

	* dbus/dbus-transport-unix.c: a number of modifications to cope
	with removing messages_pending
	(check_write_watch): properly handle
	DBUS_AUTH_STATE_WAITING_FOR_MEMORY; adapt to removal of
	messages_pending stuff
	(check_read_watch): properly handle WAITING_FOR_MEMORY and
	AUTHENTICATED cases
	(unix_handle_watch): after writing, see if the write watch can be
	removed
	(unix_do_iteration): assert that write_watch/read_watch are
	non-NULL rather than testing that they aren't, since they
	aren't allowed to be NULL. check_write_watch() at the end so
	we add the watch if we did not finish writing (e.g. got EAGAIN)

	* dbus/dbus-transport-protected.h: remove messages_pending call,
	since it resulted in too much inefficient watch adding/removing;
	instead we now require that the transport user does an iteration
	after queueing outgoing messages, and after trying the first
	write() we add a write watch if we got EAGAIN or exceeded our
	max bytes to write per iteration setting

	* dbus/dbus-string.c (_dbus_string_validate_signature): add this
	function

	* dbus/dbus-server-unix.c (unix_finalize): the socket name was
	freed and then accessed, valgrind flagged this bug, fix it

	* dbus/dbus-message.c: fix several bugs where HEADER_FIELD_LAST was taken
	as the last valid field plus 1, where really it is equal to the
	last valid field. Corrects some message corruption issues.

	* dbus/dbus-mainloop.c: verbosity changes

	* dbus/dbus-keyring.c (_dbus_keyring_new_homedir): handle OOM
	instead of aborting in one of the test codepaths

	* dbus/dbus-internals.c (_dbus_verbose_real): fix a bug that
	caused not printing the pid ever again if a verbose was missing
	the newline at the end
	(_dbus_header_field_to_string): add HEADER_FIELD_SIGNATURE

	* dbus/dbus-connection.c: verbosity changes;
	(dbus_connection_has_messages_to_send): new function
	(_dbus_connection_message_sent): no longer call transport->messages_pending
	(_dbus_connection_send_preallocated_unlocked): do one iteration to
	try to write() immediately, so we can avoid the write watch. This
	is the core purpose of this patchset
	(_dbus_connection_get_dispatch_status_unlocked): if disconnected,
	dump the outgoing message queue, so nobody will get confused
	trying to send them or thinking stuff is pending to be sent

	* bus/test.c: verbosity changes

	* bus/driver.c: verbosity/assertion changes

	* bus/dispatch.c: a bunch of little tweaks to get it working again
	because this patchset changes when/where you need to block.
2004-11-26 01:53:13 +00:00
bus 2004-11-25 Havoc Pennington <hp@redhat.com> 2004-11-26 01:53:13 +00:00
dbus 2004-11-25 Havoc Pennington <hp@redhat.com> 2004-11-26 01:53:13 +00:00
doc 2004-11-25 Havoc Pennington <hp@redhat.com> 2004-11-26 01:53:13 +00:00
gcj 2003-06-23 Anders Carlsson <andersca@codefactory.se> 2003-06-23 17:39:48 +00:00
glib 2004-11-13 Havoc Pennington <hp@redhat.com> 2004-11-13 07:07:47 +00:00
mono Patch from Johan Fischer to make mono doc install create the doc dir. 2004-11-02 10:19:20 +00:00
python 2004-09-16 David Zeuthen <david@fubar.dk> 2004-09-16 19:56:26 +00:00
qt 2004-10-29 Colin Walters <walters@redhat.com> 2004-10-29 19:20:02 +00:00
test 2004-11-25 Havoc Pennington <hp@redhat.com> 2004-11-26 01:53:13 +00:00
tools Add comment about workaround. 2004-10-29 18:52:30 +00:00
.cvsignore 2003-02-16 Havoc Pennington <hp@pobox.com> 2003-02-16 07:20:54 +00:00
acinclude.m4 2003-09-21 Seth Nickell <seth@gnome.org> 2003-09-22 05:45:59 +00:00
AUTHORS 2004-10-21 Colin Walters <walters@verbum.org> 2004-10-22 02:14:00 +00:00
autogen.sh 2003-05-15 Havoc Pennington <hp@redhat.com> 2003-05-15 19:59:19 +00:00
ChangeLog 2004-11-25 Havoc Pennington <hp@redhat.com> 2004-11-26 01:53:13 +00:00
configure.in 2004-10-29 Colin Walters <walters@redhat.com> 2004-10-29 19:17:57 +00:00
COPYING 2004-08-09 Havoc Pennington <hp@redhat.com> 2004-08-10 03:07:01 +00:00
dbus-1.pc.in 2003-04-29 Havoc Pennington <hp@redhat.com> 2003-04-29 21:56:37 +00:00
dbus-glib-1.pc.in 2003-06-22 Anders Carlsson <andersca@codefactory.se> 2003-06-22 06:53:42 +00:00
dbus-sharp.pc.in Remove glib-sharp from Libs flag. 2004-06-10 12:55:28 +00:00
Doxyfile.in 2004-06-02 Kristian Høgsberg <krh@redhat.com> 2004-06-02 13:13:14 +00:00
HACKING fix address to mail about release 2004-08-12 23:06:30 +00:00
INSTALL initial import of "dbus" skeleton 2002-11-21 16:41:33 +00:00
Makefile.am 2004-07-24 Havoc Pennington <hp@redhat.com> 2004-07-25 03:52:48 +00:00
Makefile.cvs Match kde schematics 2003-11-23 08:07:04 +00:00
NEWS 2004-08-12 Havoc Pennington <hp@redhat.com> 2004-08-12 23:02:41 +00:00
README add a couple of notes about libdbus vs. bindings 2004-08-10 02:18:37 +00:00

D-BUS is a simple IPC library based on messages.

See also the file HACKING for notes of interest to developers working on D-BUS.

See http://www.freedesktop.org/software/dbus/ for lots of documentation, 
mailing lists, etc.

Note
===

A core concept of the D-BUS implementation is that "libdbus" is
intended to be a low-level API, similar to Xlib. Most programmers are
intended to use the bindings to GLib, Qt, Python, Mono, Java, or
whatever. These bindings have varying levels of completeness.

Configuration flags
===

These are the dbus-specific configuration flags that can be given to
the ./configure program.

  --enable-qt      enable Qt-friendly client library
  --enable-glib    enable GLib-friendly client library
  --enable-mono    enable mono bindings
  --enable-mono-docs build mono documentation (requires monodoc)
  --enable-tests   enable unit test code
  --enable-ansi    enable -ansi -pedantic gcc flags
  --enable-verbose-mode support verbose debug mode
  --enable-asserts include assertion checks
  --enable-checks  include sanity checks on public API
  --enable-docs    build documentation (requires Doxygen and jade)
  --enable-gcov    compile with coverage profiling instrumentation (gcc only)
  --enable-python  build python bindings (reqires Pyrex >= 0.9)

  --with-xml=libxml/expat           XML library to use
  --with-init-scripts=redhat        Style of init scripts to install
  --with-session-socket-dir=dirname Where to put sockets for the per-login-session message bus
  --with-test-socket-dir=dirname    Where to put sockets for make check
  --with-system-pid-file=pidfile    PID file for systemwide daemon
  --with-system-socket=filename     UNIX domain socket for systemwide daemon


API/ABI Policy
===

D-BUS API/ABI and protocol necessarily remain in flux until we are
sure it will meet the various needs it's intended to meet. This means
we need to see some significant sample usage in the contexts of GNOME,
KDE, desktop applications, and systemwide uses such as print queue
monitoring, hotplug events, or whatever. We need the flexibility to
incorporate feedback from this sample usage.

Once we feel confident in the protocol and the API, we will release a 
version 1.0. At that point, the intent is:

 - The protocol will never be broken again; any message bus should 
   work with any client forever. However, extensions are possible
   where the protocol is extensible.

 - If the library API is modified incompatibly, we will rename it 
   as in http://ometer.com/parallel.html - in other words, 
   it will always be possible to compile against and use the older 
   API, and apps will always get the API they expect.

Until 1.0 is released, feedback that requires API changes may be
incorporated into D-BUS. This may break the API, the ABI, the
protocol, or all three.

To avoid a huge soname, the plan is to increment the soname only
between official stable releases, not with every development snapshot.
Versions numbered 0.x are considered development snapshots.

Until 1.0 is released, you have to define -DDBUS_API_SUBJECT_TO_CHANGE
just as a safety check to be sure everyone is aware of this API/ABI
policy and has the right expectations.

We do need people to test the APIs, so please do use the development
snapshots of D-BUS. They are intended to work and we do actively
address bugs.

However, if you're shipping a commercial binary-only application that
needs to keep running on M future versions of N operating systems, you
might want to include your own copy of D-BUS rather than relying on
the installed copy, for example.