hmm, make check is currently not passing.
* doc/dbus-specification.xml: add requirement that custom type
names follow the same rules as interface names.
* dbus/dbus-protocol.h: change some of the byte codes, to avoid
duplication and allow 'c' to be 'custom'; dict is now 'm' for
'map'
* doc/dbus-specification.xml: update type codes to match
dbus-protocol.h, using the ASCII byte values. Rename type NAMED to
CUSTOM. Add type OBJECT_PATH to the spec.
2003-10-17 Havoc Pennington <hp@redhat.com>
* bus/driver.c (create_unique_client_name): use "." as separator
in base service names instead of '-'
* dbus/dbus-string.c (_dbus_string_get_byte): allow getting nul
byte at the end of the string
* dbus/dbus-internals.h (_DBUS_LIKELY, _DBUS_UNLIKELY): add
optimization macros since string validation seems to be a slow
point.
* doc/dbus-specification.xml: restrict valid
service/interface/member/error names. Add test suite code for the
name validation.
* dbus/dbus-string.c: limit service/interface/member/error names
to [0-9][A-Z][a-z]_
* dbus/dbus-connection.c (dbus_connection_dispatch): add missing
format arg to verbose spew
* glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): if not out of
memory, return instead of g_error
* test/test-service.c (path_message_func): support emitting a
signal on request
* dbus/dbus-bus.c (init_connections_unlocked): only fill in
activation bus type if DBUS_BUS_ACTIVATION was set; default to
assuming the activation bus was the session bus so that services
started manually will still register.
(init_connections_unlocked): fix so that in OOM situation we get
the same semantics when retrying the function
* test/test-service.c (main): change to use path registration, to
test those codepaths; register with DBUS_BUS_ACTIVATION rather
than DBUS_BUS_SESSION
* glib/dbus-gtest-main.c: bracket with #ifdef DBUS_BUILD_TESTS
* Makefile.am (GCOV_DIRS): remove "test", we don't care about test
coverage of the tests
(coverage-report.txt): don't move the .da and .bbg files around
* bus/connection.c (bus_pending_reply_expired): either cancel or
execute, not both
(bus_connections_check_reply): use unlink, not remove_link, as we
don't want to free the link; fixes double free mess
* dbus/dbus-pending-call.c (dbus_pending_call_block): fix in case
where no reply was received
* dbus/dbus-connection.c (_dbus_pending_call_complete_and_unlock):
fix a refcount leak
* bus/signals.c (match_rule_matches): add special cases for the
bus driver, so you can match on sender/destination for it.
* dbus/dbus-sysdeps.c (_dbus_abort): print backtrace if
DBUS_PRINT_BACKTRACE is set
* dbus/dbus-internals.c: add pid to assertion failure messages
* dbus/dbus-connection.c: add message type code to the debug spew
* glib/dbus-gproxy.c (gproxy_get_match_rule): match rules want
sender=foo not service=foo
* dbus/dbus-bus.c (dbus_bus_get): if the activation bus is the
session bus but DBUS_SESSION_BUS_ADDRESS isn't set, use
DBUS_ACTIVATION_ADDRESS instead
* bus/activation.c: set DBUS_SESSION_BUS_ADDRESS,
DBUS_SYSTEM_BUS_ADDRESS if appropriate
* bus/bus.c (bus_context_new): handle OOM copying bus type into
context struct
* dbus/dbus-message.c (dbus_message_iter_get_object_path): new function
(dbus_message_iter_get_object_path_array): new function (half
finished, disabled for the moment)
* glib/dbus-gproxy.c (dbus_gproxy_end_call): properly handle
DBUS_MESSAGE_TYPE_ERROR
* tools/dbus-launch.c (babysit): support DBUS_DEBUG_OUTPUT to
avoid redirecting stderr to /dev/null
(babysit): close stdin if not doing the "exit_with_session" thing
* dbus/dbus-sysdeps.c (_dbus_become_daemon): delete some leftover
debug code; change DBUS_DEBUG_OUTPUT to only enable stderr, not
stdout/stdin, so things don't get confused
* bus/system.conf.in: fix to allow replies, I modified .conf
instead of .conf.in again.
Added test code that 1) starts an actual bus daemon and 2) uses
DBusGProxy; fixed bugs that were revealed by the test. Lots
more testing possible, but this is the basic framework.
* glib/dbus-gproxy.c (dbus_gproxy_manager_unregister): remove
empty proxy lists from the proxy list hash
* dbus/dbus-message.c (dbus_message_iter_get_args_valist): add a
couple of return_if_fail checks
* dbus/dbus-pending-call.c (_dbus_pending_call_new): use dbus_new0
to allocate, so everything is cleared to NULL as it should be.
* glib/dbus-gmain.c (dbus_connection_setup_with_g_main): pass
source as data to dbus_connection_set_timeout_functions() as the
timeout functions expected
* test/glib/run-test.sh: add a little script to start up a message
bus and run tests using it
* tools/dbus-launch.1: updates
* tools/dbus-launch.c (main): add --config-file option
* tools/dbus-launch.c (main): remove confusing else if (runprog)
that could never be reached.
* dbus/dbus-message.c (dbus_message_new_method_return)
(dbus_message_new_error, dbus_message_new_signal): set the
no-reply-expected flag on all these. Redundant, but may
as well be consistent.
* glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): rename from
dbus_gproxy_oneway_call
* glib/dbus-gmain.c (dbus_connection_setup_with_g_main)
(dbus_server_setup_with_g_main): fix to allow calling them more
than once on the same args
(dbus_bus_get_with_g_main): new function
* Merge dbus-object-names branch. To see the entire patch
do cvs diff -r DBUS_OBJECT_NAMES_BRANCHPOINT -r dbus-object-names,
it's huuuuge though.
To revert, I tagged DBUS_BEFORE_OBJECT_NAMES_MERGE.
2003-09-28 Havoc Pennington <hp@pobox.com>
* HACKING: update to reflect new server
2003-09-26 Seth Nickell <seth@gnome.org>
* python/dbus.py:
* python/examples/example-signals.py:
Start implementing some notions of signals. The API
is really terrible, but they sort of work (with the
exception of being able to filter by service, and to
transmit signals *as* a particular service). Need to
figure out how to make messages come from the service
we registered :-(
* python/dbus_bindings.pyx.in:
Removed duplicate message_handler callbacks.
2003-09-25 Havoc Pennington <hp@redhat.com>
* bus/session.conf.in: fix my mess
2003-09-25 Havoc Pennington <hp@pobox.com>
* bus/session.conf.in: fix security policy, reported by Seth Nickell
2003-09-25 Seth Nickell <seth@gnome.org>
* python/examples/example-service.py:
Johan notices complete wrong code in example-service, but
completely wrong in a way that works exactly the same (!).
Johan is confused, how could this possibly work? Example
code fails to serve purpose of making things clear.
Seth fixes.
2003-09-25 Mark McLoughlin <mark@skynet.ie>
* doc/dbus-specification.sgml: don't require header fields
to be 4-byte aligned and specify that fields should be
distinguished from padding by the fact that zero is not
a valid field name.
* doc/TODO: remove re-alignment item and add item to doc
the OBJECT_PATH type.
* dbus/dbus-message.c:
(HeaderField): rename the original member to value_offset
and introduce a name_offset member to keep track of where
the field actually begins.
(adjust_field_offsets): remove.
(append_int_field), (append_uint_field),
(append_string_field): don't align the start of the header
field to a 4-byte boundary.
(get_next_field): impl finding the next marhsalled field
after a given field.
(re_align_field_recurse): impl re-aligning a number of
already marshalled fields.
(delete_field): impl deleting a field of any type and
re-aligning any following fields.
(delete_int_or_uint_field), (delete_string_field): remove.
(set_int_field), (set_uint_field): no need to re-check
that we have the correct type for the field.
(set_string_field): ditto and impl re-aligning any
following fields.
(decode_header_data): update to take into account that
the fields aren't 4-byte aligned any more and the new
way to distinguish padding from header fields. Also,
don't exit when there is too much header padding.
(process_test_subdir): print the directory.
(_dbus_message_test): add test to make sure a following
field is re-aligned correctly after field deletion.
* dbus/dbus-string.[ch]:
(_dbus_string_insert_bytes): rename from insert_byte and
allow the insert of multiple bytes.
(_dbus_string_test): test inserting multiple bytes.
* dbus/dbus-marshal.c: (_dbus_marshal_set_string): add
warning note to docs about having to re-align any
marshalled values following the string.
* dbus/dbus-message-builder.c:
(append_string_field), (_dbus_message_data_load):
don't align the header field.
* dbus/dbus-auth.c: (process_test_subdir): print the
directory.
* test/break-loader.c: (randomly_add_one_byte): upd. for
insert_byte change.
* test/data/invalid-messages/bad-header-field-alignment.message:
new test case.
* test/data/valid-messages/unknown-header-field.message: shove
a dict in the unknown field.
2003-09-25 Seth Nickell <seth@gnome.org>
* python/dbus.py:
* python/dbus_bindings.pyx.in:
Handle return values.
* python/examples/example-client.py:
* python/examples/example-service.py:
Pass back return values from the service to the client.
2003-09-24 Seth Nickell <seth@gnome.org>
* python/dbus.py:
Connect Object methods (when you are sharing an object) up... pass
in a list of methods to be shared. Sharing all the methods just
worked out too weird. You can now create nice Services over the
DBus in Python. :-)
* python/dbus_bindings.pyx.in:
Keep references to user_data tuples passed into C functions so
Python doesn't garbage collect on us.
Implement MethodReturn and Error subclasses of Message for creating
DBusMessage's of those types.
* python/examples/example-client.py:
* python/examples/example-service.py:
Simple example code showing both how create DBus services and objects,
and how to use them.
2003-09-23 Havoc Pennington <hp@pobox.com>
* glib/dbus-gproxy.c (dbus_gproxy_manager_filter): implement
2003-09-23 Havoc Pennington <hp@redhat.com>
* glib/dbus-gproxy.c (dbus_gproxy_connect_signal): implement
(dbus_gproxy_disconnect_signal): implement
(dbus_gproxy_manager_remove_signal_match): implement
(dbus_gproxy_manager_add_signal_match): implement
(dbus_gproxy_oneway_call): implement
2003-09-23 Havoc Pennington <hp@pobox.com>
* glib/dbus-gproxy.c (struct DBusGProxy): convert to a GObject
subclass. This means dropping the transparent thread safety of the
proxy; you now need a separate proxy per-thread, or your own
locking on the proxy. Probably right anyway.
(dbus_gproxy_ref, dbus_gproxy_unref): nuke, just use g_object_ref
2003-09-22 Havoc Pennington <hp@redhat.com>
* glib/dbus-gproxy.c (dbus_gproxy_manager_get): implement
2003-09-21 Seth Nickell <seth@gnome.org>
First checkin of the Python bindings.
* python/.cvsignore:
* python/Makefile.am:
* python/dbus_bindings.pyx.in:
* python/dbus_h_wrapper.h:
Pieces for Pyrex to operate on, building a dbus_bindings.so
python module for low-level access to the DBus APIs.
* python/dbus.py:
High-level Python module for accessing DBus objects.
* configure.in:
* Makefile.am:
Build stuff for the python bindings.
* acinclude.m4:
Extra macro needed for finding the Python C header files.
2003-09-21 Havoc Pennington <hp@pobox.com>
* glib/dbus-gproxy.c (dbus_gproxy_manager_new): start
implementing the proxy manager, didn't get very far.
* dbus/dbus-bus.c (dbus_bus_add_match): new
(dbus_bus_remove_match): new
* glib/dbus-gproxy.c (dbus_gproxy_new_for_service): add a
path_name argument; adjust the other not-yet-implemented
gproxy constructors to be what I think they should be.
2003-09-21 Havoc Pennington <hp@pobox.com>
* dbus/dbus-bus.c (dbus_bus_get): set exit_on_disconnect to TRUE
by default for message bus connections.
* dbus/dbus-connection.c (dbus_connection_dispatch): exit if
exit_on_disconnect flag is set and we process the disconnected
signal.
(dbus_connection_set_exit_on_disconnect): new function
2003-09-21 Havoc Pennington <hp@pobox.com>
Get matching rules mostly working in the bus; only actually
parsing the rule text remains. However, the client side of
"signal connections" hasn't been started, this patch is only the
bus side.
* dbus/dispatch.c: fix for the matching rules changes
* bus/driver.c (bus_driver_handle_remove_match)
(bus_driver_handle_add_match): send an ack reply from these
method calls
* glib/dbus-gproxy.c (dbus_gproxy_begin_call): fix order of
arguments, reported by Seth Nickell
* bus/config-parser.c (append_rule_from_element): support
eavesdrop=true|false attribute on policies so match rules
can be prevented from snooping on the system bus.
* bus/dbus-daemon-1.1.in: consistently use terminology "sender"
and "destination" in attribute names; fix some docs bugs;
add eavesdrop=true|false attribute
* bus/driver.c (bus_driver_handle_add_match)
(bus_driver_handle_remove_match): handle AddMatch, RemoveMatch
messages
* dbus/dbus-protocol.h (DBUS_SERVICE_ORG_FREEDESKTOP_BROADCAST): get
rid of broadcast service concept, signals are just always broadcast
* bus/signals.c, bus/dispatch.c, bus/connection.c, bus/bus.c:
mostly implement matching rules stuff (currently only exposed as signal
connections)
2003-09-21 Mark McLoughlin <mark@skynet.ie>
* doc/dbus-specification.sgml: Change the header field name
to be an enum and update the rest of the spec to reference
the fields using the conventinal name.
* dbus/dbus-protocol.h: update to reflect the spec.
* doc/TODO: add item to remove the 4 byte alignment requirement.
* dbus/dbus-message.c: Remove the code to generalise the
header/body length and serial number header fields as named
header fields so we can reference field names using the
protocol values.
(append_int_field), (append_uint_field), (append_string_field):
Append the field name as a byte rather than four chars.
(delete_int_or_uint_field), (delete_string_field): reflect the
fact that the field name and typecode now occupy 4 bytes instead
of 8.
(decode_string_field), (decode_header_data): update to reflect
protocol changes and move the field specific encoding from
decode_string_field() back into decode_header_data().
* dbus/dbus-internals.[ch]: (_dbus_header_field_to_string):
Add utility to aid debugging.
* dbus/dbus-message-builder.c:
(append_string_field), (_dbus_message_data_load): Update to
reflect protocol changes; Change the FIELD_NAME directive
to HEADER_FIELD and allow it to take the field's conventional
name rather than the actual value.
* test/data/*/*.message: Update to use HEADER_FIELD instead
of FIELD_NAME; Always align the header on an 8 byte boundary
*before* updating the header length.
2003-09-15 Havoc Pennington <hp@pobox.com>
* dbus/dbus-pending-call.c: add the get/set object data
boilerplate as for DBusConnection, etc. Use generic object data
for the notify callback.
* glib/dbus-gparser.c (parse_node): parse child nodes
* tools/dbus-viewer.c: more hacking on the dbus-viewer
* glib/dbus-gutils.c (_dbus_gutils_split_path): add a file to
contain functions shared between the convenience lib and the
installed lib
* glib/Makefile.am (libdbus_glib_1_la_LDFLAGS): add
-export-symbols-regex to the GLib library
* dbus/dbus-object-tree.c (_dbus_object_tree_dispatch_and_unlock):
fix the locking in here, and add a default handler for
Introspect() that just returns sub-nodes.
2003-09-14 Havoc Pennington <hp@pobox.com>
* glib/dbus-gthread.c (dbus_g_thread_init): rename to make g_foo
rather than gfoo consistent
* glib/dbus-gproxy.h: delete for now, move contents to
dbus-glib.h, because the include files don't work right since we
aren't in the dbus/ subdir.
* glib/dbus-gproxy.c (dbus_gproxy_send): finish implementing
(dbus_gproxy_end_call): finish
(dbus_gproxy_begin_call): finish
* glib/dbus-gmain.c (dbus_set_g_error): new
* glib/dbus-gobject.c (handle_introspect): include information
about child nodes in the introspection
* dbus/dbus-connection.c (dbus_connection_list_registered): new
function to help in implementation of introspection
* dbus/dbus-object-tree.c
(_dbus_object_tree_list_registered_and_unlock): new function
2003-09-12 Havoc Pennington <hp@pobox.com>
* glib/dbus-gidl.h: add common base class for all the foo_info
types
* tools/dbus-viewer.c: add GTK-based introspection UI thingy
similar to kdcop
* test/Makefile.am: try test srcdir -ef . in addition to test
srcdir = ., one of them should work (yeah lame)
* glib/Makefile.am: build the "idl" parser stuff as a convenience
library
* glib/dbus-gparser.h: make description_load routines return
NodeInfo* not Parser*
* Makefile.am (SUBDIRS): build test dir after all library dirs
* configure.in: add GTK+ detection
2003-09-07 Havoc Pennington <hp@pobox.com>
* Make Doxygen contented.
2003-09-07 Havoc Pennington <hp@pobox.com>
* doc/dbus-specification.sgml: more updates
2003-09-06 Havoc Pennington <hp@pobox.com>
* doc/dbus-specification.sgml: partial updates
* bus/dbus-daemon-1.1.in: fix the config file docs for the
zillionth time; hopefully I edited the right file this time.
* bus/config-parser.c (append_rule_from_element): support
send_type, send_path, receive_type, receive_path
* bus/policy.c: add message type and path to the list of things
that can be "firewalled"
2003-09-06 Havoc Pennington <hp@pobox.com>
* dbus/dbus-connection.c (dbus_connection_register_fallback): add this
(dbus_connection_register_object_path): make this not handle
messages to paths below the given path
2003-09-03 Havoc Pennington <hp@pobox.com>
* test/glib/Makefile.am: add this with random glib-linked test
programs
* glib/Makefile.am: remove the random test programs from here,
leave only the unit tests
* glib/dbus-gobject.c (_dbus_gobject_test): add test for
uscore/javacaps conversion, and fix
(get_object_property, set_object_property): change to .NET
convention for mapping props to methods, set_FooBar/get_FooBar,
since one language has such a convention we may as well copy it.
Plus real methods in either getFooBar or get_foo_bar style won't
collide with this convention.
2003-09-01 Havoc Pennington <hp@pobox.com>
* glib/dbus-gparser.c: implement
* glib/dbus-gobject.c: start implementing skeletons support
* configure.in: when disabling checks/assert, also define
G_DISABLE_ASSERT and G_DISABLE_CHECKS
2003-09-01 Havoc Pennington <hp@pobox.com>
* glib/Makefile.am: rearrange a bunch of files and get "make
check" framework set up
2003-08-31 Havoc Pennington <hp@pobox.com>
* fix build with --disable-tests
2003-08-30 Havoc Pennington <hp@pobox.com>
* dbus/dbus-connection.c: purge DBusMessageHandler
* dbus/dbus-message-handler.c: remove DBusMessageHandler, just
use callbacks everywhere
2003-08-30 Havoc Pennington <hp@pobox.com>
* test/data/valid-config-files/system.d/test.conf: change to
root for the user so warnings don't get printed
* dbus/dbus-message.c: add dbus_message_get_path,
dbus_message_set_path
* dbus/dbus-object-tree.c (do_test_dispatch): add test of
dispatching to a path
* dbus/dbus-string.c (_dbus_string_validate_path): add
* dbus/dbus-marshal.c (_dbus_demarshal_object_path): implement
(_dbus_marshal_object_path): implement
* dbus/dbus-protocol.h (DBUS_HEADER_FIELD_PATH): new header field
to contain the path to the target object
(DBUS_HEADER_FIELD_SENDER_SERVICE): rename
DBUS_HEADER_FIELD_SENDER to explicitly say it's the sender service
2003-08-30 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object-tree.c: write tests and fix the discovered bugs
2003-08-29 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object-tree.c: modify to allow overlapping paths to be
registered
(struct DBusObjectSubtree): shrink this
a lot, since we may have a lot of them
(_dbus_object_tree_free_all_unlocked): implement
(_dbus_object_tree_dispatch_and_unlock): implement
2003-08-29 Havoc Pennington <hp@pobox.com>
* dbus/dbus-internals.h: fix _DBUS_N_GLOBAL_LOCKS
2003-08-28 Havoc Pennington <hp@pobox.com>
purge DBusObjectID
* dbus/dbus-connection.c: port to no ObjectID, create a
DBusObjectTree, rename ObjectTree to ObjectPath in public API
* dbus/dbus-connection.h (struct DBusObjectTreeVTable): delete
everything except UnregisterFunction and MessageFunction
* dbus/dbus-marshal.c: port away from DBusObjectID,
add DBUS_TYPE_OBJECT_PATH
* dbus/dbus-object-registry.[hc], dbus/dbus-object.[hc],
dbus/dbus-objectid.[hc]: remove these, we are moving to
path-based object IDs
2003-08-25 Havoc Pennington <hp@pobox.com>
Just noticed that dbus_message_test is hosed, I wonder when I
broke that. I thought make check was passing earlier...
* dbus/dbus-object-tree.c: add new "object tree" to match DCOP
container tree, will replace most of dbus-object-registry
* dbus/dbus-string.c (_dbus_string_append_printf_valist): fix C99
screwup
2003-08-19 Havoc Pennington <hp@pobox.com>
* dbus/dbus-message.c (decode_string_field): support FIELD_SENDER
(dbus_message_is_error): fix this function
* bus/dbus-daemon-1.1: clarify logic on when <deny>/<allow> rules
match
* bus/policy.c (bus_client_policy_check_can_receive): fix code to
reflect clarified man page
(bus_client_policy_check_can_send): ditto
* bus/session.conf.in: fixup
* bus/system.conf.in: fixup
2003-08-18 Havoc Pennington <hp@redhat.com>
* dbus/dbus-hash.c (_dbus_hash_table_insert_two_strings): fix
* dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
dumb bug created earlier (wrong order of args to
decode_header_data())
* tools/dbus-send.c: port
* tools/dbus-print-message.c (print_message): port
* test/data/*messages: port all messages over
* dbus/dbus-message-builder.c: support including
message type
* bus/driver.c: port over
* bus/dispatch.c: port over to new stuff
* dbus/dbus-connection.c (_dbus_connection_new_for_transport):
rename disconnect signal to "Disconnected"
2003-08-17 Havoc Pennington <hp@pobox.com>
This doesn't compile yet, but syncing up so I can hack on it from
work. What are branches for if not broken code? ;-)
* dbus/dbus-protocol.h: remove DBUS_HEADER_FIELD_NAME, add
DBUS_HEADER_FIELD_INTERFACE, DBUS_HEADER_FIELD_MEMBER,
DBUS_HEADER_FIELD_ERROR_NAME
* dbus/dbus-hash.c: Introduce DBUS_HASH_TWO_STRINGS as hack to use
for the interface+member pairs
(string_hash): change to use g_str_hash algorithm
(find_direct_function, find_string_function): refactor these to
share most code.
* dbus/dbus-message.c: port all of this over to support
interface/member fields instead of name field
* dbus/dbus-object-registry.c: port over
* dbus/dbus-string.c (_dbus_string_validate_interface): rename
from _dbus_string_validate_name
* bus/dbus-daemon-1.1: change file format for the
<deny>/<allow> stuff to match new message naming scheme
* bus/policy.c: port over
* bus/config-parser.c: parse new format
2003-08-16 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object-registry.c (add_and_remove_objects): remove
broken assertion
* glib/dbus-gproxy.c: some hacking
2003-08-15 Havoc Pennington <hp@redhat.com>
* dbus/dbus-pending-call.c (dbus_pending_call_block): implement
* dbus/dbus-connection.c
(dbus_connection_send_with_reply_and_block): factor out internals;
change to convert any error replies to DBusError instead of
returning them as a message
2003-08-15 Havoc Pennington <hp@pobox.com>
* dbus/dbus-connection.c,
dbus/dbus-pending-call.c: Finish the pending call stuff
2003-08-14 Havoc Pennington <hp@redhat.com>
* dbus/dbus-pending-call.c: start on new object that will replace
DBusMessageHandler and ReplyHandlerData for tracking outstanding
replies
* dbus/dbus-gproxy.c: start on proxy object used to communicate
with remote interfaces
* dbus/dbus-gidl.c: do the boring boilerplate in here
2003-08-12 Havoc Pennington <hp@pobox.com>
* bus/dispatch.c (bus_dispatch): make this return proper
DBusHandlerResult to avoid DBUS_ERROR_UNKNOWN_METHOD
* dbus/dbus-errors.c (dbus_set_error): use
_dbus_string_append_printf_valist
* dbus/dbus-string.c (_dbus_string_append_printf_valist)
(_dbus_string_append_printf): new
* dbus/dbus-errors.h (DBUS_ERROR_UNKNOWN_MESSAGE): change to
UNKNOWN_METHOD
* dbus/dbus-connection.c (dbus_connection_dispatch): handle
DBUS_HANDLER_RESULT_NEED_MEMORY; send default error reply if a
message is unhandled.
2003-08-11 Havoc Pennington <hp@pobox.com>
* bus/test.c (client_disconnect_handler): change to return
HANDLED (would have been REMOVE_MESSAGE)
* dbus/dbus-object.h (enum DBusHandlerResult): rename to
HANDLED/NOT_YET_HANDLED instead of
REMOVE_MESSAGE/ALLOW_MORE_HANDLERS to make it clearer how it
should be used.
2003-08-10 Havoc Pennington <hp@pobox.com>
* tools/dbus-send.c (main): add --type argument, for now
supporting only method_call and signal types.
* tools/dbus-print-message.c: print message type
* dbus/dbus-connection.c (_dbus_connection_new_for_transport):
init connection->objects
* doc/dbus-specification.sgml: fix sgml
* bus/*.c: port over to object-instance API changes
* test/test-service.c: ditto
* dbus/dbus-message.c (dbus_message_create_header): allow #NULL
name, we will have to fix up the rest of the code to also handle
this
(dbus_message_new): generic message-creation call
(set_string_field): allow appending name field
2003-08-06 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object-registry.c: implement signal connection
and dispatch
* dbus/dbus-connection.c (_dbus_connection_unref_unlocked): new
* dbus/dbus-internals.c (_dbus_memdup): new function
2003-08-02 Havoc Pennington <hp@pobox.com>
* dbus/dbus-message.c (dbus_message_get_no_reply)
(dbus_message_set_no_reply): add these and remove
set_is_error/get_is_error
* dbus/dbus-protocol.h, doc/dbus-specification.sgml:
remove the ERROR flag, since there's now an ERROR type
2003-08-01 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object-registry.c (_dbus_object_registry_handle_and_unlock):
implement
* dbus/dbus-message.c (dbus_message_get_type): new function
* doc/dbus-specification.sgml: add "type" byte to messages
2003-08-01 Havoc Pennington <hp@pobox.com>
* dbus/dbus-protocol.h (DBUS_MESSAGE_TYPE_*): introduce
a message type enum to distinguish kinds of message
(DBUS_HEADER_FLAG_NO_REPLY_EXPECTED): flag for a message
that need not be replied to
2003-08-01 Havoc Pennington <hp@pobox.com>
* dbus/dbus-marshal.c: adapt to DBusObjectID changes
(unpack_8_octets): fix no-64-bit-int bug
* dbus/dbus-object-registry.c (validate_id): validate the
connection ID bits, not just the instance ID.
* dbus/dbus-connection.c (_dbus_connection_init_id): initialize
the connection-global 33 bits of the object ID
* dbus/dbus-object-registry.c (info_from_entry): fill in
object ID in the new way
* dbus/dbus-objectid.h: rather than high/low bits, specifically
define server/client/instance bits.
2003-07-30 Havoc Pennington <hp@pobox.com>
* dbus/dbus-connection.c (dbus_connection_register_object): fix
build
2003-07-13 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object.h (struct DBusObjectVTable): add padding
fields to DBusObjectVTable and DBusObjectInfo
2003-07-12 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object-registry.c: implement unit test,
fix bugs discovered in process
* dbus/dbus-connection.c: remove handler_table and
register_handler(), add DBusObjectRegistry usage
* dbus/dbus-objectid.c (dbus_object_id_is_null)
(dbus_object_id_set_null): new functions
2003-07-08 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object.c: implement some of this
* dbus/dbus-object-registry.c
(_dbus_object_registry_add_and_unlock): fill in the object_id out
param
(_dbus_object_registry_new): handle OOM
2003-07-08 Havoc Pennington <hp@pobox.com>
* dbus/dbus-object.h: sketch out an API for registering objects
with a connection, that allows us to use as little as 24 bytes
per object and lets application code represent an object in
any conceivable way.
* dbus/dbus-object-registry.c: implement the hard bits of the
DBusConnection aspect of object API. Not yet wired up.
2003-07-06 Havoc Pennington <hp@pobox.com>
* dbus/dbus-marshal.c (_dbus_marshal_set_object_id): new function
(_dbus_marshal_object_id): new
(_dbus_demarshal_object_id): new
(_dbus_marshal_get_arg_end_pos): support object ID type, and
consolidate identical switch cases. Don't conditionalize handling
of DBUS_TYPE_UINT64, need to handle the type always.
(_dbus_marshal_validate_arg): consolidate identical cases, and
handle DBUS_TYPE_OBJECT_ID
* dbus/dbus-objectid.c: new file with DBusObjectID data type.
* dbus/dbus-protocol.h: add DBUS_TYPE_OBJECT_ID
2003-09-28 Havoc Pennington <hp@pobox.com>
* real 0.13 release
2003-09-28 Havoc Pennington <hp@pobox.com>
* doc/Makefile.am (dbus-specification.html): testing a funky hack
to work with Debian db2html
2003-09-28 Havoc Pennington <hp@pobox.com>
* configure.in: 0.13
* doc/Makefile.am (dbus-test-plan.html): accept nonexistence of
stylesheet-images for benefit of Debian
Change back to using filesystem-linked sockets for the system
bus, so only root can create the default system bus address.
* bus/system.conf.in: change to use
DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
* dbus/Makefile.am (INCLUDES): remove DBUS_SYSTEM_BUS_PATH define
from here.
* configure.in: define DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
here, and AC_DEFINE DBUS_SYSTEM_PATH
2003-08-09 Anders Carlsson <andersca@codefactory.se>
* doc/TODO:
* doc/busconfig.dtd:
Add busconfig DTD.
2003-08-09 Anders Carlsson <andersca@codefactory.se>
* doc/dbus-specification.sgml:
Add activation reply values.
2003-08-05 Havoc Pennington <hp@redhat.com>
* configure.in: 0.12
2003-08-05 Anders Carlsson <andersca@codefactory.se>
* glib/dbus-gmain.c: (watch_fd_new), (watch_fd_ref),
(watch_fd_unref), (dbus_gsource_check), (dbus_gsource_dispatch),
(add_watch), (remove_watch), (create_source):
Refcount fds, fixes some reentrancy issues.
2003-07-30 Havoc Pennington <hp@redhat.com>
* dbus/dbus-bus.c (init_connections_unlocked): fix default system
bus address to be abstract if we have abstract sockets
* NEWS: update
2003-07-28 Havoc Pennington <hp@redhat.com>
* bus/messagebus.in: fix to avoid processname/servicename
confusion, from Michael Kearey
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=100965
2003-07-23 Havoc Pennington <hp@pobox.com>
* dbus/dbus-message.c (dbus_message_iter_get_named):
fix from Andy Hanton to remove broken "+1"
2003-07-16 Havoc Pennington <hp@pobox.com>
* tools/dbus-launch.c (babysit): close stdout/stderr in the
babysitter process, as suggested by Thomas Leonard, so
an "eval `dbus-launch --exit-with-session`" will actually
return
2003-07-16 Havoc Pennington <hp@pobox.com>
* configure.in: print out EXPANDED_* variables in the summary at
the end; clean up the code that computes EXPANDED_ variables and
get the ones using exec_prefix right. Should make things work
when you build without --prefix
* glib/dbus-gproxy.c (struct DBusGProxy): convert to a GObject
subclass. This means dropping the transparent thread safety of the
proxy; you now need a separate proxy per-thread, or your own
locking on the proxy. Probably right anyway.
(dbus_gproxy_ref, dbus_gproxy_unref): nuke, just use g_object_ref
* glib/dbus-gproxy.c (dbus_gproxy_manager_new): start
implementing the proxy manager, didn't get very far.
* dbus/dbus-bus.c (dbus_bus_add_match): new
(dbus_bus_remove_match): new
* glib/dbus-gproxy.c (dbus_gproxy_new_for_service): add a
path_name argument; adjust the other not-yet-implemented
gproxy constructors to be what I think they should be.
Get matching rules mostly working in the bus; only actually
parsing the rule text remains. However, the client side of
"signal connections" hasn't been started, this patch is only the
bus side.
* dbus/dispatch.c: fix for the matching rules changes
* bus/driver.c (bus_driver_handle_remove_match)
(bus_driver_handle_add_match): send an ack reply from these
method calls
* glib/dbus-gproxy.c (dbus_gproxy_begin_call): fix order of
arguments, reported by Seth Nickell
* bus/config-parser.c (append_rule_from_element): support
eavesdrop=true|false attribute on policies so match rules
can be prevented from snooping on the system bus.
* bus/dbus-daemon-1.1.in: consistently use terminology "sender"
and "destination" in attribute names; fix some docs bugs;
add eavesdrop=true|false attribute
* bus/driver.c (bus_driver_handle_add_match)
(bus_driver_handle_remove_match): handle AddMatch, RemoveMatch
messages
* dbus/dbus-protocol.h (DBUS_SERVICE_ORG_FREEDESKTOP_BROADCAST): get
rid of broadcast service concept, signals are just always broadcast
* bus/signals.c, bus/dispatch.c, bus/connection.c, bus/bus.c:
mostly implement matching rules stuff (currently only exposed as signal
connections)
* dbus/dbus-pending-call.c: add the get/set object data
boilerplate as for DBusConnection, etc. Use generic object data
for the notify callback.
* glib/dbus-gparser.c (parse_node): parse child nodes
* tools/dbus-viewer.c: more hacking on the dbus-viewer
* glib/dbus-gutils.c (_dbus_gutils_split_path): add a file to
contain functions shared between the convenience lib and the
installed lib
* glib/Makefile.am (libdbus_glib_1_la_LDFLAGS): add
-export-symbols-regex to the GLib library
* dbus/dbus-object-tree.c (_dbus_object_tree_dispatch_and_unlock):
fix the locking in here, and add a default handler for
Introspect() that just returns sub-nodes.
2003-09-14 Havoc Pennington <hp@pobox.com>
* glib/dbus-gthread.c (dbus_g_thread_init): rename to make g_foo
rather than gfoo consistent
* glib/dbus-gproxy.h: delete for now, move contents to
dbus-glib.h, because the include files don't work right since we
aren't in the dbus/ subdir.
* glib/dbus-gproxy.c (dbus_gproxy_send): finish implementing
(dbus_gproxy_end_call): finish
(dbus_gproxy_begin_call): finish
* glib/dbus-gmain.c (dbus_set_g_error): new
* glib/dbus-gobject.c (handle_introspect): include information
about child nodes in the introspection
* dbus/dbus-connection.c (dbus_connection_list_registered): new
function to help in implementation of introspection
* dbus/dbus-object-tree.c
(_dbus_object_tree_list_registered_and_unlock): new function
2003-09-12 Havoc Pennington <hp@pobox.com>
* glib/dbus-gidl.h: add common base class for all the foo_info
types
* tools/dbus-viewer.c: add GTK-based introspection UI thingy
similar to kdcop
* test/Makefile.am: try test srcdir -ef . in addition to test
srcdir = ., one of them should work (yeah lame)
* glib/Makefile.am: build the "idl" parser stuff as a convenience
library
* glib/dbus-gparser.h: make description_load routines return
NodeInfo* not Parser*
* Makefile.am (SUBDIRS): build test dir after all library dirs
* configure.in: add GTK+ detection
* test/glib/Makefile.am: add this with random glib-linked test
programs
* glib/Makefile.am: remove the random test programs from here,
leave only the unit tests
* glib/dbus-gobject.c (_dbus_gobject_test): add test for
uscore/javacaps conversion, and fix
(get_object_property, set_object_property): change to .NET
convention for mapping props to methods, set_FooBar/get_FooBar,
since one language has such a convention we may as well copy it.
Plus real methods in either getFooBar or get_foo_bar style won't
collide with this convention.
* glib/dbus-gparser.c: implement
* glib/dbus-gobject.c: start implementing skeletons support
* configure.in: when disabling checks/assert, also define
G_DISABLE_ASSERT and G_DISABLE_CHECKS
* test/data/valid-config-files/system.d/test.conf: change to
root for the user so warnings don't get printed
* dbus/dbus-message.c: add dbus_message_get_path,
dbus_message_set_path
* dbus/dbus-object-tree.c (do_test_dispatch): add test of
dispatching to a path
* dbus/dbus-string.c (_dbus_string_validate_path): add
* dbus/dbus-marshal.c (_dbus_demarshal_object_path): implement
(_dbus_marshal_object_path): implement
* dbus/dbus-protocol.h (DBUS_HEADER_FIELD_PATH): new header field
to contain the path to the target object
(DBUS_HEADER_FIELD_SENDER_SERVICE): rename
DBUS_HEADER_FIELD_SENDER to explicitly say it's the sender service
* dbus/dbus-object-tree.c: modify to allow overlapping paths to be
registered
(struct DBusObjectSubtree): shrink this
a lot, since we may have a lot of them
(_dbus_object_tree_free_all_unlocked): implement
(_dbus_object_tree_dispatch_and_unlock): implement
purge DBusObjectID
* dbus/dbus-connection.c: port to no ObjectID, create a
DBusObjectTree, rename ObjectTree to ObjectPath in public API
* dbus/dbus-connection.h (struct DBusObjectTreeVTable): delete
everything except UnregisterFunction and MessageFunction
* dbus/dbus-marshal.c: port away from DBusObjectID,
add DBUS_TYPE_OBJECT_PATH
* dbus/dbus-object-registry.[hc], dbus/dbus-object.[hc],
dbus/dbus-objectid.[hc]: remove these, we are moving to
path-based object IDs
* dbus/dbus-hash.c (_dbus_hash_table_insert_two_strings): fix
* dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
dumb bug created earlier (wrong order of args to
decode_header_data())
* tools/dbus-send.c: port
* tools/dbus-print-message.c (print_message): port
* test/data/*messages: port all messages over
* dbus/dbus-message-builder.c: support including
message type
* bus/driver.c: port over
* bus/dispatch.c: port over to new stuff
* dbus/dbus-connection.c (_dbus_connection_new_for_transport):
rename disconnect signal to "Disconnected"
* dbus/dbus-pending-call.c (dbus_pending_call_block): implement
* dbus/dbus-connection.c
(dbus_connection_send_with_reply_and_block): factor out internals;
change to convert any error replies to DBusError instead of
returning them as a message
* dbus/dbus-pending-call.c: start on new object that will replace
DBusMessageHandler and ReplyHandlerData for tracking outstanding
replies
* dbus/dbus-gproxy.c: start on proxy object used to communicate
with remote interfaces
* dbus/dbus-gidl.c: do the boring boilerplate in here
* bus/test.c (client_disconnect_handler): change to return
HANDLED (would have been REMOVE_MESSAGE)
* dbus/dbus-object.h (enum DBusHandlerResult): rename to
HANDLED/NOT_YET_HANDLED instead of
REMOVE_MESSAGE/ALLOW_MORE_HANDLERS to make it clearer how it
should be used.
* tools/dbus-send.c (main): add --type argument, for now
supporting only method_call and signal types.
* tools/dbus-print-message.c: print message type
* dbus/dbus-connection.c (_dbus_connection_new_for_transport):
init connection->objects
* doc/dbus-specification.sgml: fix sgml
* bus/*.c: port over to object-instance API changes
* test/test-service.c: ditto
* dbus/dbus-message.c (dbus_message_create_header): allow #NULL
name, we will have to fix up the rest of the code to also handle
this
(dbus_message_new): generic message-creation call
(set_string_field): allow appending name field
* dbus/dbus-dataslot.c (_dbus_data_slot_allocator_unref)
(_dbus_data_slot_allocator_alloc): rework these to keep a
reference count on each slot and automatically manage a global
slot ID variable passed in by address
* bus/bus.c: convert to new dataslot API
* dbus/dbus-bus.c: convert to new dataslot API
* dbus/dbus-connection.c: convert to new dataslot API
* dbus/dbus-server.c: convert to new dataslot API
* glib/dbus-gmain.c: ditto
* bus/test.c: ditto
* bus/connection.c: ditto
* bus/config-parser.c (merge_included): merge in policies from
child configuration file.
* bus/policy.c (bus_policy_merge): function to merge two policies
together
* dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix to avoid
calling _dbus_marshal_validate_arg() for every byte in a byte
array, etc.
* dbus/dbus-message-handler.c: use atomic reference counting to
reduce number of locks slightly; the global lock in here sucks
* dbus/dbus-connection.c
(_dbus_connection_update_dispatch_status_and_unlock): variant of
update_dispatch_status that can be called with lock held; then use
in a couple places to reduce locking/unlocking
(dbus_connection_send): hold the lock over the whole function
instead of acquiring it twice.
* dbus/dbus-timeout.c (_dbus_timeout_new): handle OOM
* bus/connection.c (bus_connections_setup_connection): fix access
to already-freed memory.
* dbus/dbus-connection.c: keep a little cache of linked list
nodes, to avoid using the global linked list alloc lock in the
normal send-message case. Instead we just use the connection lock
that we already have to take.
* dbus/dbus-list.c (_dbus_list_find_last): new function
* dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec):
change to use a struct for the atomic type; fix docs,
they return value before increment, not after increment.
* dbus/dbus-string.c (_dbus_string_append_4_aligned)
(_dbus_string_append_8_aligned): new functions to try to
microoptimize this operation.
(reallocate_for_length): break this out of set_length(), to
improve profile info, and also so we can consider inlining the
set_length() part.
* dbus/dbus-message.c (dbus_message_new_empty_header): init data
strings with some preallocation, cuts down on our calls to realloc
a fair bit. Though if we can get the "move entire string to empty
string" optimization below to kick in here, it would be better.
* dbus/dbus-string.c (_dbus_string_move): just call
_dbus_string_move_len
(_dbus_string_move_len): add a special case for moving
an entire string into an empty string; we can just
swap the string data instead of doing any reallocs.
(_dbus_string_init_preallocated): new function
Write a "test-profile" that does echo client-server with threads;
profile reveals lock contention, memcpy/realloc of buffers, and
UTF-8 validation as hot spots. 20% of lock contention eliminated
with dbus_atomic_inc/dec implementation on x86. Much remaining
contention is global mempool locks for GList and DBusList.
* dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec): add
x86 implementation
* dbus/dbus-connection.c (struct DBusConnection): use
dbus_atomic_t for the reference count
* dbus/dbus-message.c (struct DBusMessage): declare
dbus_atomic_t values as volatile
* configure.in: code to detect ability to use atomic integer
operations in assembly, from GLib patch
* dbus/dbus-internals.c (_dbus_verbose_real): call getpid every
time, tired of it being wrong in threads and forked processes
* glib/test-profile.c: a little program to bounce messages back
and forth between threads and eat CPU
* dbus/dbus-connection.c: add debug spew macros for debugging
thread locks; include config.h at top; fix deadlock in
dbus_connection_flush()
* glib/dbus-gmain.c: docs cleanups
* dbus/dbus-types.h: add docs on int64 types
* dbus/dbus-memory.c: fix docs to avoid putting private API in
public API docs section
* glib/dbus-gmain.c: adapt to watch changes
* bus/bus.c, bus/activation.c, etc.: adjust to watch changes
* dbus/dbus-server.h: remove dbus_server_handle_watch
* dbus/dbus-connection.h: remove dbus_connection_handle_watch
* dbus/dbus-watch.c (dbus_watch_handle): change DBusWatch to work
like DBusTimeout, so we don't need dbus_connection_handle_watch
etc.
* bus/.cvsignore:
* glib/.cvsignore:
* test/.cvsignore:
Added files to cvsignore
* dbus/dbus-message.h:
* dbus/dbus-message.c: (dbus_message_iter_get_named):
Make get_named() take two out argument and return a boolean.
(dbus_message_iter_get_args_valist):
Update usage of get_named().
(dbus_message_iter_append_byte):
Fix typo
(dbus_message_iter_append_named)
Fix typo
(message_iter_test), (check_message_handling_type), (_dbus_message_test):
More tests.
Implemented recursive types, named types and new-style iters
* bus/driver.c:
* glib/test-thread-client.c: (thread_func):
* glib/test-thread-server.c: (handle_test_message):
* test/test-service.c: (handle_echo):
Update to new api
* dbus/Makefile.am:
* dbus/dbus-dict.c:
* dbus/dbus-dict.h:
* dbus/dbus.h
Remove DBusDict
* dbus/dbus-internals.c: (_dbus_type_to_string):
Update for new types.
* dbus/dbus-marshal.[ch]:
Implement recursive types and the new marshalling format.
Remove hardcoded dict marshalling.
Marshal named types.
* dbus/dbus-message-builder.c:
Add BYTE_ARRAY.
Remove references to old types
* dbus/dbus-message.[ch]:
New non-refcounted iter API that supports recursive iters.
Use iters for appending, including support for recursive
iters.
Add byte and named type support.
Update everything to new marshalling formats.
Add tests for new API.
* dbus/dbus-protocol.h:
Remove old array types.
Add types: BYTE, ARRAY, DICT, NAMED
* dbus/dbus-string.c:
* dbus/dbus-sysdeps.c:
Make parse_double locale safe.
* dbus/dbus-test-main.c:
Call setlocale.
* dbus/dbus-test.c:
Kill dict test
* doc/dbus-specification.sgml:
Update spec
* test/data/incomplete-messages/missing-body.message:
* test/data/invalid-messages/bad-boolean.message:
* test/data/invalid-messages/bad-boolean-array.message:
* test/data/invalid-messages/boolean-array-length-too-long.message-raw:
* test/data/invalid-messages/boolean-has-no-value.message-raw:
* test/data/invalid-messages/too-short-dict.message:
* test/data/valid-messages/dict-simple.message:
* test/data/valid-messages/dict.message:
* test/data/valid-messages/emptiness.message:
* test/data/valid-messages/lots-of-arguments.message:
* test/data/valid-messages/no-padding.message:
* test/data/valid-messages/recursive-types.message:
Add missing NAME fields
Fix up dicts & arrays
* test/data/invalid-messages/dict-with-nil-value.message:
Removed, this is not invalid anymore.
* test/data/valid-messages/recursive-types.message:
Add new test for deeply recursive types.
* dbus/dbus-sysdeps.c (_dbus_set_fd_nonblocking): move to this
file
* dbus/dbus-errors.c (dbus_set_error, dbus_set_error_const): allow
NULL argument for "message" if the error is a well-known one,
fill in a generic message in this case.
* dbus/dbus-errors.h (DBusResultCode): Kill DBusResultCode in
favor of DBusError
* bus/test.c (bus_test_flush_bus): add
* bus/policy.c (bus_policy_test): test code stub
* dbus/dbus-string.c (_dbus_string_validate_utf8): oops, unbreak
this. always run the test suite before commit...
* bus/*: adapt to DBusConnection API changes
* glib/dbus-gmain.c: adapt to DBusConnection API changes,
requires renaming stuff to avoid dbus_connection_dispatch name
conflict.
* dbus/dbus-transport.c (_dbus_transport_queue_messages): new
function
* dbus/dbus-message.c (_dbus_message_loader_queue_messages):
separate from _dbus_message_loader_return_buffer()
* dbus/dbus-connection.c (dbus_connection_get_n_messages): remove
this, because it's now always broken to use; the number of
messages in queue vs. the number still buffered by the message
loader is undefined/meaningless. Should use
dbus_connection_get_dispatch_state().
(dbus_connection_dispatch): rename from
dbus_connection_dispatch_message
Make it pass the Hello handling test including all OOM codepaths.
Now to do other messages...
* bus/services.c (bus_service_remove_owner): fix crash when
removing owner from an empty list of owners
(bus_registry_ensure): don't leave service in the list of
a connection's owned services if we fail to put the service
in the hash table.
* bus/connection.c (bus_connection_preallocate_oom_error): set
error flag on the OOM error.
* dbus/dbus-connection.c (_dbus_connection_new_for_transport):
handle _dbus_transport_set_connection failure
* dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd): modify
to create watches up front and simply enable/disable them as
needed.
(unix_connection_set): this can now fail on OOM
* dbus/dbus-timeout.c, dbus/dbus-watch.c: add concept
of enabling/disabling a watch or timeout.
* bus/loop.c (bus_loop_iterate): don't touch disabled
watches/timeouts
* glib/dbus-gmain.c: adapt to enable/disable watches and timeouts
