1 2003-10-11 Havoc Pennington <hp@pobox.com>
3 * dbus/dbus-message.c (_dbus_message_test): add more test
4 coverage, but #if 0 for now since they uncover a bug
5 not fixed yet; I think in re_align_field_recurse()
6 (re_align_field_recurse): add FIXME about broken assertion
8 * dbus/dbus-sysdeps.c (_dbus_sysdeps_test): add more test coverage
10 * bus/connection.c: share a couple code bits with expirelist.c
12 * bus/expirelist.h, bus/expirelist.c: implement a generic
13 expire-items-after-N-seconds facility, was going to share between
14 expiring connections and replies, decided not to use for expiring
17 * COPYING: include AFL 2.0 (still need to change all the file headers)
19 2003-10-09 Havoc Pennington <hp@redhat.com>
21 * configure.in: define DBUS_HAVE_GCC33_GCOV if we have
22 gcc 3.3. Not that we do anything about it yet.
24 * bus/signals.c (bus_match_rule_parse): impose max length on the
27 * dbus/dbus-protocol.h: add DBUS_MAXIMUM_MATCH_RULE_LENGTH
29 2003-10-09 Havoc Pennington <hp@redhat.com>
31 Make matching rules theoretically work (add parser).
33 * bus/bus.c (bus_context_check_security_policy): fix up to handle
34 the case where destination is explicitly specified as bus driver
35 and someone else is eavesdropping.
37 * bus/policy.c (bus_client_policy_check_can_receive): fix up
38 definition of eavesdropping and assertion
40 * tools/dbus-send.c (main): use dbus_message_type_from_string
42 * bus/signals.c (bus_match_rule_parse): implement
44 * dbus/dbus-message.c (dbus_message_type_from_string): new
46 * dbus/dbus-errors.h (DBUS_ERROR_MATCH_RULE_INVALID): add
48 2003-10-02 Havoc Pennington <hp@pobox.com>
50 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): rename from
51 dbus_gproxy_oneway_call
53 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main)
54 (dbus_server_setup_with_g_main): fix to allow calling them more
55 than once on the same args
56 (dbus_bus_get_with_g_main): new function
58 2003-10-02 Havoc Pennington <hp@redhat.com>
60 * doc/dbus-tutorial.xml: write some stuff
62 2003-09-29 Havoc Pennington <hp@pobox.com>
64 * configure.in: split checks for Doxygen from XML docs, check for
67 * doc/Makefile.am: XML-ify all the docs, and add a blank
70 2003-09-29 Havoc Pennington <hp@pobox.com>
72 * Merge dbus-object-names branch. To see the entire patch
73 do cvs diff -r DBUS_OBJECT_NAMES_BRANCHPOINT -r dbus-object-names,
75 To revert, I tagged DBUS_BEFORE_OBJECT_NAMES_MERGE.
77 2003-09-28 Havoc Pennington <hp@pobox.com>
79 * HACKING: update to reflect new server
81 2003-09-26 Seth Nickell <seth@gnome.org>
84 * python/examples/example-signals.py:
86 Start implementing some notions of signals. The API
87 is really terrible, but they sort of work (with the
88 exception of being able to filter by service, and to
89 transmit signals *as* a particular service). Need to
90 figure out how to make messages come from the service
93 * python/dbus_bindings.pyx.in:
95 Removed duplicate message_handler callbacks.
97 2003-09-25 Havoc Pennington <hp@redhat.com>
99 * bus/session.conf.in: fix my mess
101 2003-09-25 Havoc Pennington <hp@pobox.com>
103 * bus/session.conf.in: fix security policy, reported by Seth Nickell
105 2003-09-25 Seth Nickell <seth@gnome.org>
107 * python/examples/example-service.py:
109 Johan notices complete wrong code in example-service, but
110 completely wrong in a way that works exactly the same (!).
111 Johan is confused, how could this possibly work? Example
112 code fails to serve purpose of making things clear.
115 2003-09-25 Mark McLoughlin <mark@skynet.ie>
117 * doc/dbus-specification.sgml: don't require header fields
118 to be 4-byte aligned and specify that fields should be
119 distinguished from padding by the fact that zero is not
122 * doc/TODO: remove re-alignment item and add item to doc
123 the OBJECT_PATH type.
125 * dbus/dbus-message.c:
126 (HeaderField): rename the original member to value_offset
127 and introduce a name_offset member to keep track of where
128 the field actually begins.
129 (adjust_field_offsets): remove.
130 (append_int_field), (append_uint_field),
131 (append_string_field): don't align the start of the header
132 field to a 4-byte boundary.
133 (get_next_field): impl finding the next marhsalled field
135 (re_align_field_recurse): impl re-aligning a number of
136 already marshalled fields.
137 (delete_field): impl deleting a field of any type and
138 re-aligning any following fields.
139 (delete_int_or_uint_field), (delete_string_field): remove.
140 (set_int_field), (set_uint_field): no need to re-check
141 that we have the correct type for the field.
142 (set_string_field): ditto and impl re-aligning any
144 (decode_header_data): update to take into account that
145 the fields aren't 4-byte aligned any more and the new
146 way to distinguish padding from header fields. Also,
147 don't exit when there is too much header padding.
148 (process_test_subdir): print the directory.
149 (_dbus_message_test): add test to make sure a following
150 field is re-aligned correctly after field deletion.
152 * dbus/dbus-string.[ch]:
153 (_dbus_string_insert_bytes): rename from insert_byte and
154 allow the insert of multiple bytes.
155 (_dbus_string_test): test inserting multiple bytes.
157 * dbus/dbus-marshal.c: (_dbus_marshal_set_string): add
158 warning note to docs about having to re-align any
159 marshalled values following the string.
161 * dbus/dbus-message-builder.c:
162 (append_string_field), (_dbus_message_data_load):
163 don't align the header field.
165 * dbus/dbus-auth.c: (process_test_subdir): print the
168 * test/break-loader.c: (randomly_add_one_byte): upd. for
171 * test/data/invalid-messages/bad-header-field-alignment.message:
174 * test/data/valid-messages/unknown-header-field.message: shove
175 a dict in the unknown field.
177 2003-09-25 Seth Nickell <seth@gnome.org>
180 * python/dbus_bindings.pyx.in:
182 Handle return values.
184 * python/examples/example-client.py:
185 * python/examples/example-service.py:
187 Pass back return values from the service to the client.
189 2003-09-24 Seth Nickell <seth@gnome.org>
193 Connect Object methods (when you are sharing an object) up... pass
194 in a list of methods to be shared. Sharing all the methods just
195 worked out too weird. You can now create nice Services over the
198 * python/dbus_bindings.pyx.in:
200 Keep references to user_data tuples passed into C functions so
201 Python doesn't garbage collect on us.
203 Implement MethodReturn and Error subclasses of Message for creating
204 DBusMessage's of those types.
206 * python/examples/example-client.py:
207 * python/examples/example-service.py:
209 Simple example code showing both how create DBus services and objects,
212 2003-09-23 Havoc Pennington <hp@pobox.com>
214 * glib/dbus-gproxy.c (dbus_gproxy_manager_filter): implement
216 2003-09-23 Havoc Pennington <hp@redhat.com>
218 * glib/dbus-gproxy.c (dbus_gproxy_connect_signal): implement
219 (dbus_gproxy_disconnect_signal): implement
220 (dbus_gproxy_manager_remove_signal_match): implement
221 (dbus_gproxy_manager_add_signal_match): implement
222 (dbus_gproxy_oneway_call): implement
224 2003-09-23 Havoc Pennington <hp@pobox.com>
226 * glib/dbus-gproxy.c (struct DBusGProxy): convert to a GObject
227 subclass. This means dropping the transparent thread safety of the
228 proxy; you now need a separate proxy per-thread, or your own
229 locking on the proxy. Probably right anyway.
230 (dbus_gproxy_ref, dbus_gproxy_unref): nuke, just use g_object_ref
232 2003-09-22 Havoc Pennington <hp@redhat.com>
234 * glib/dbus-gproxy.c (dbus_gproxy_manager_get): implement
236 2003-09-21 Seth Nickell <seth@gnome.org>
238 First checkin of the Python bindings.
241 * python/Makefile.am:
242 * python/dbus_bindings.pyx.in:
243 * python/dbus_h_wrapper.h:
245 Pieces for Pyrex to operate on, building a dbus_bindings.so
246 python module for low-level access to the DBus APIs.
250 High-level Python module for accessing DBus objects.
255 Build stuff for the python bindings.
259 Extra macro needed for finding the Python C header files.
261 2003-09-21 Havoc Pennington <hp@pobox.com>
263 * glib/dbus-gproxy.c (dbus_gproxy_manager_new): start
264 implementing the proxy manager, didn't get very far.
266 * dbus/dbus-bus.c (dbus_bus_add_match): new
267 (dbus_bus_remove_match): new
269 * glib/dbus-gproxy.c (dbus_gproxy_new_for_service): add a
270 path_name argument; adjust the other not-yet-implemented
271 gproxy constructors to be what I think they should be.
273 2003-09-21 Havoc Pennington <hp@pobox.com>
275 * dbus/dbus-bus.c (dbus_bus_get): set exit_on_disconnect to TRUE
276 by default for message bus connections.
278 * dbus/dbus-connection.c (dbus_connection_dispatch): exit if
279 exit_on_disconnect flag is set and we process the disconnected
281 (dbus_connection_set_exit_on_disconnect): new function
283 2003-09-21 Havoc Pennington <hp@pobox.com>
285 Get matching rules mostly working in the bus; only actually
286 parsing the rule text remains. However, the client side of
287 "signal connections" hasn't been started, this patch is only the
290 * dbus/dispatch.c: fix for the matching rules changes
292 * bus/driver.c (bus_driver_handle_remove_match)
293 (bus_driver_handle_add_match): send an ack reply from these
296 * glib/dbus-gproxy.c (dbus_gproxy_begin_call): fix order of
297 arguments, reported by Seth Nickell
299 * bus/config-parser.c (append_rule_from_element): support
300 eavesdrop=true|false attribute on policies so match rules
301 can be prevented from snooping on the system bus.
303 * bus/dbus-daemon-1.1.in: consistently use terminology "sender"
304 and "destination" in attribute names; fix some docs bugs;
305 add eavesdrop=true|false attribute
307 * bus/driver.c (bus_driver_handle_add_match)
308 (bus_driver_handle_remove_match): handle AddMatch, RemoveMatch
311 * dbus/dbus-protocol.h (DBUS_SERVICE_ORG_FREEDESKTOP_BROADCAST): get
312 rid of broadcast service concept, signals are just always broadcast
314 * bus/signals.c, bus/dispatch.c, bus/connection.c, bus/bus.c:
315 mostly implement matching rules stuff (currently only exposed as signal
318 2003-09-21 Mark McLoughlin <mark@skynet.ie>
320 * doc/dbus-specification.sgml: Change the header field name
321 to be an enum and update the rest of the spec to reference
322 the fields using the conventinal name.
324 * dbus/dbus-protocol.h: update to reflect the spec.
326 * doc/TODO: add item to remove the 4 byte alignment requirement.
328 * dbus/dbus-message.c: Remove the code to generalise the
329 header/body length and serial number header fields as named
330 header fields so we can reference field names using the
332 (append_int_field), (append_uint_field), (append_string_field):
333 Append the field name as a byte rather than four chars.
334 (delete_int_or_uint_field), (delete_string_field): reflect the
335 fact that the field name and typecode now occupy 4 bytes instead
337 (decode_string_field), (decode_header_data): update to reflect
338 protocol changes and move the field specific encoding from
339 decode_string_field() back into decode_header_data().
341 * dbus/dbus-internals.[ch]: (_dbus_header_field_to_string):
342 Add utility to aid debugging.
344 * dbus/dbus-message-builder.c:
345 (append_string_field), (_dbus_message_data_load): Update to
346 reflect protocol changes; Change the FIELD_NAME directive
347 to HEADER_FIELD and allow it to take the field's conventional
348 name rather than the actual value.
350 * test/data/*/*.message: Update to use HEADER_FIELD instead
351 of FIELD_NAME; Always align the header on an 8 byte boundary
352 *before* updating the header length.
354 2003-09-15 Havoc Pennington <hp@pobox.com>
356 * dbus/dbus-pending-call.c: add the get/set object data
357 boilerplate as for DBusConnection, etc. Use generic object data
358 for the notify callback.
360 * glib/dbus-gparser.c (parse_node): parse child nodes
362 * tools/dbus-viewer.c: more hacking on the dbus-viewer
364 * glib/dbus-gutils.c (_dbus_gutils_split_path): add a file to
365 contain functions shared between the convenience lib and the
368 * glib/Makefile.am (libdbus_glib_1_la_LDFLAGS): add
369 -export-symbols-regex to the GLib library
371 * dbus/dbus-object-tree.c (_dbus_object_tree_dispatch_and_unlock):
372 fix the locking in here, and add a default handler for
373 Introspect() that just returns sub-nodes.
375 2003-09-14 Havoc Pennington <hp@pobox.com>
377 * glib/dbus-gthread.c (dbus_g_thread_init): rename to make g_foo
378 rather than gfoo consistent
380 * glib/dbus-gproxy.h: delete for now, move contents to
381 dbus-glib.h, because the include files don't work right since we
382 aren't in the dbus/ subdir.
384 * glib/dbus-gproxy.c (dbus_gproxy_send): finish implementing
385 (dbus_gproxy_end_call): finish
386 (dbus_gproxy_begin_call): finish
388 * glib/dbus-gmain.c (dbus_set_g_error): new
390 * glib/dbus-gobject.c (handle_introspect): include information
391 about child nodes in the introspection
393 * dbus/dbus-connection.c (dbus_connection_list_registered): new
394 function to help in implementation of introspection
396 * dbus/dbus-object-tree.c
397 (_dbus_object_tree_list_registered_and_unlock): new function
399 2003-09-12 Havoc Pennington <hp@pobox.com>
401 * glib/dbus-gidl.h: add common base class for all the foo_info
404 * tools/dbus-viewer.c: add GTK-based introspection UI thingy
407 * test/Makefile.am: try test srcdir -ef . in addition to test
408 srcdir = ., one of them should work (yeah lame)
410 * glib/Makefile.am: build the "idl" parser stuff as a convenience
413 * glib/dbus-gparser.h: make description_load routines return
414 NodeInfo* not Parser*
416 * Makefile.am (SUBDIRS): build test dir after all library dirs
418 * configure.in: add GTK+ detection
420 2003-09-07 Havoc Pennington <hp@pobox.com>
422 * Make Doxygen contented.
424 2003-09-07 Havoc Pennington <hp@pobox.com>
426 * doc/dbus-specification.sgml: more updates
428 2003-09-06 Havoc Pennington <hp@pobox.com>
430 * doc/dbus-specification.sgml: partial updates
432 * bus/dbus-daemon-1.1.in: fix the config file docs for the
433 zillionth time; hopefully I edited the right file this time.
435 * bus/config-parser.c (append_rule_from_element): support
436 send_type, send_path, receive_type, receive_path
438 * bus/policy.c: add message type and path to the list of things
439 that can be "firewalled"
441 2003-09-06 Havoc Pennington <hp@pobox.com>
443 * dbus/dbus-connection.c (dbus_connection_register_fallback): add this
444 (dbus_connection_register_object_path): make this not handle
445 messages to paths below the given path
447 2003-09-03 Havoc Pennington <hp@pobox.com>
449 * test/glib/Makefile.am: add this with random glib-linked test
452 * glib/Makefile.am: remove the random test programs from here,
453 leave only the unit tests
455 * glib/dbus-gobject.c (_dbus_gobject_test): add test for
456 uscore/javacaps conversion, and fix
457 (get_object_property, set_object_property): change to .NET
458 convention for mapping props to methods, set_FooBar/get_FooBar,
459 since one language has such a convention we may as well copy it.
460 Plus real methods in either getFooBar or get_foo_bar style won't
461 collide with this convention.
463 2003-09-01 Havoc Pennington <hp@pobox.com>
465 * glib/dbus-gparser.c: implement
467 * glib/dbus-gobject.c: start implementing skeletons support
469 * configure.in: when disabling checks/assert, also define
470 G_DISABLE_ASSERT and G_DISABLE_CHECKS
472 2003-09-01 Havoc Pennington <hp@pobox.com>
474 * glib/Makefile.am: rearrange a bunch of files and get "make
475 check" framework set up
477 2003-08-31 Havoc Pennington <hp@pobox.com>
479 * fix build with --disable-tests
481 2003-08-30 Havoc Pennington <hp@pobox.com>
483 * dbus/dbus-connection.c: purge DBusMessageHandler
485 * dbus/dbus-message-handler.c: remove DBusMessageHandler, just
486 use callbacks everywhere
488 2003-08-30 Havoc Pennington <hp@pobox.com>
490 * test/data/valid-config-files/system.d/test.conf: change to
491 root for the user so warnings don't get printed
493 * dbus/dbus-message.c: add dbus_message_get_path,
494 dbus_message_set_path
496 * dbus/dbus-object-tree.c (do_test_dispatch): add test of
497 dispatching to a path
499 * dbus/dbus-string.c (_dbus_string_validate_path): add
501 * dbus/dbus-marshal.c (_dbus_demarshal_object_path): implement
502 (_dbus_marshal_object_path): implement
504 * dbus/dbus-protocol.h (DBUS_HEADER_FIELD_PATH): new header field
505 to contain the path to the target object
506 (DBUS_HEADER_FIELD_SENDER_SERVICE): rename
507 DBUS_HEADER_FIELD_SENDER to explicitly say it's the sender service
509 2003-08-30 Havoc Pennington <hp@pobox.com>
511 * dbus/dbus-object-tree.c: write tests and fix the discovered bugs
513 2003-08-29 Havoc Pennington <hp@pobox.com>
515 * dbus/dbus-object-tree.c: modify to allow overlapping paths to be
517 (struct DBusObjectSubtree): shrink this
518 a lot, since we may have a lot of them
519 (_dbus_object_tree_free_all_unlocked): implement
520 (_dbus_object_tree_dispatch_and_unlock): implement
522 2003-08-29 Havoc Pennington <hp@pobox.com>
524 * dbus/dbus-internals.h: fix _DBUS_N_GLOBAL_LOCKS
526 2003-08-28 Havoc Pennington <hp@pobox.com>
530 * dbus/dbus-connection.c: port to no ObjectID, create a
531 DBusObjectTree, rename ObjectTree to ObjectPath in public API
533 * dbus/dbus-connection.h (struct DBusObjectTreeVTable): delete
534 everything except UnregisterFunction and MessageFunction
536 * dbus/dbus-marshal.c: port away from DBusObjectID,
537 add DBUS_TYPE_OBJECT_PATH
539 * dbus/dbus-object-registry.[hc], dbus/dbus-object.[hc],
540 dbus/dbus-objectid.[hc]: remove these, we are moving to
541 path-based object IDs
543 2003-08-25 Havoc Pennington <hp@pobox.com>
545 Just noticed that dbus_message_test is hosed, I wonder when I
546 broke that. I thought make check was passing earlier...
548 * dbus/dbus-object-tree.c: add new "object tree" to match DCOP
549 container tree, will replace most of dbus-object-registry
551 * dbus/dbus-string.c (_dbus_string_append_printf_valist): fix C99
554 2003-08-19 Havoc Pennington <hp@pobox.com>
556 * dbus/dbus-message.c (decode_string_field): support FIELD_SENDER
557 (dbus_message_is_error): fix this function
559 * bus/dbus-daemon-1.1: clarify logic on when <deny>/<allow> rules
562 * bus/policy.c (bus_client_policy_check_can_receive): fix code to
563 reflect clarified man page
564 (bus_client_policy_check_can_send): ditto
566 * bus/session.conf.in: fixup
568 * bus/system.conf.in: fixup
570 2003-08-18 Havoc Pennington <hp@redhat.com>
572 * dbus/dbus-hash.c (_dbus_hash_table_insert_two_strings): fix
574 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
575 dumb bug created earlier (wrong order of args to
576 decode_header_data())
578 * tools/dbus-send.c: port
580 * tools/dbus-print-message.c (print_message): port
582 * test/data/*messages: port all messages over
584 * dbus/dbus-message-builder.c: support including
587 * bus/driver.c: port over
589 * bus/dispatch.c: port over to new stuff
591 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
592 rename disconnect signal to "Disconnected"
594 2003-08-17 Havoc Pennington <hp@pobox.com>
596 This doesn't compile yet, but syncing up so I can hack on it from
597 work. What are branches for if not broken code? ;-)
599 * dbus/dbus-protocol.h: remove DBUS_HEADER_FIELD_NAME, add
600 DBUS_HEADER_FIELD_INTERFACE, DBUS_HEADER_FIELD_MEMBER,
601 DBUS_HEADER_FIELD_ERROR_NAME
603 * dbus/dbus-hash.c: Introduce DBUS_HASH_TWO_STRINGS as hack to use
604 for the interface+member pairs
605 (string_hash): change to use g_str_hash algorithm
606 (find_direct_function, find_string_function): refactor these to
609 * dbus/dbus-message.c: port all of this over to support
610 interface/member fields instead of name field
612 * dbus/dbus-object-registry.c: port over
614 * dbus/dbus-string.c (_dbus_string_validate_interface): rename
615 from _dbus_string_validate_name
617 * bus/dbus-daemon-1.1: change file format for the
618 <deny>/<allow> stuff to match new message naming scheme
620 * bus/policy.c: port over
622 * bus/config-parser.c: parse new format
624 2003-08-16 Havoc Pennington <hp@pobox.com>
626 * dbus/dbus-object-registry.c (add_and_remove_objects): remove
629 * glib/dbus-gproxy.c: some hacking
631 2003-08-15 Havoc Pennington <hp@redhat.com>
633 * dbus/dbus-pending-call.c (dbus_pending_call_block): implement
635 * dbus/dbus-connection.c
636 (dbus_connection_send_with_reply_and_block): factor out internals;
637 change to convert any error replies to DBusError instead of
638 returning them as a message
640 2003-08-15 Havoc Pennington <hp@pobox.com>
642 * dbus/dbus-connection.c,
643 dbus/dbus-pending-call.c: Finish the pending call stuff
645 2003-08-14 Havoc Pennington <hp@redhat.com>
647 * dbus/dbus-pending-call.c: start on new object that will replace
648 DBusMessageHandler and ReplyHandlerData for tracking outstanding
651 * dbus/dbus-gproxy.c: start on proxy object used to communicate
652 with remote interfaces
654 * dbus/dbus-gidl.c: do the boring boilerplate in here
656 2003-08-12 Havoc Pennington <hp@pobox.com>
658 * bus/dispatch.c (bus_dispatch): make this return proper
659 DBusHandlerResult to avoid DBUS_ERROR_UNKNOWN_METHOD
661 * dbus/dbus-errors.c (dbus_set_error): use
662 _dbus_string_append_printf_valist
664 * dbus/dbus-string.c (_dbus_string_append_printf_valist)
665 (_dbus_string_append_printf): new
667 * dbus/dbus-errors.h (DBUS_ERROR_UNKNOWN_MESSAGE): change to
670 * dbus/dbus-connection.c (dbus_connection_dispatch): handle
671 DBUS_HANDLER_RESULT_NEED_MEMORY; send default error reply if a
672 message is unhandled.
674 2003-08-11 Havoc Pennington <hp@pobox.com>
676 * bus/test.c (client_disconnect_handler): change to return
677 HANDLED (would have been REMOVE_MESSAGE)
679 * dbus/dbus-object.h (enum DBusHandlerResult): rename to
680 HANDLED/NOT_YET_HANDLED instead of
681 REMOVE_MESSAGE/ALLOW_MORE_HANDLERS to make it clearer how it
684 2003-08-10 Havoc Pennington <hp@pobox.com>
686 * tools/dbus-send.c (main): add --type argument, for now
687 supporting only method_call and signal types.
689 * tools/dbus-print-message.c: print message type
691 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
692 init connection->objects
694 * doc/dbus-specification.sgml: fix sgml
696 * bus/*.c: port over to object-instance API changes
698 * test/test-service.c: ditto
700 * dbus/dbus-message.c (dbus_message_create_header): allow #NULL
701 name, we will have to fix up the rest of the code to also handle
703 (dbus_message_new): generic message-creation call
704 (set_string_field): allow appending name field
706 2003-08-06 Havoc Pennington <hp@pobox.com>
708 * dbus/dbus-object-registry.c: implement signal connection
711 * dbus/dbus-connection.c (_dbus_connection_unref_unlocked): new
713 * dbus/dbus-internals.c (_dbus_memdup): new function
715 2003-08-02 Havoc Pennington <hp@pobox.com>
717 * dbus/dbus-message.c (dbus_message_get_no_reply)
718 (dbus_message_set_no_reply): add these and remove
719 set_is_error/get_is_error
721 * dbus/dbus-protocol.h, doc/dbus-specification.sgml:
722 remove the ERROR flag, since there's now an ERROR type
724 2003-08-01 Havoc Pennington <hp@pobox.com>
726 * dbus/dbus-object-registry.c (_dbus_object_registry_handle_and_unlock):
729 * dbus/dbus-message.c (dbus_message_get_type): new function
731 * doc/dbus-specification.sgml: add "type" byte to messages
733 2003-08-01 Havoc Pennington <hp@pobox.com>
735 * dbus/dbus-protocol.h (DBUS_MESSAGE_TYPE_*): introduce
736 a message type enum to distinguish kinds of message
737 (DBUS_HEADER_FLAG_NO_REPLY_EXPECTED): flag for a message
738 that need not be replied to
740 2003-08-01 Havoc Pennington <hp@pobox.com>
742 * dbus/dbus-marshal.c: adapt to DBusObjectID changes
743 (unpack_8_octets): fix no-64-bit-int bug
745 * dbus/dbus-object-registry.c (validate_id): validate the
746 connection ID bits, not just the instance ID.
748 * dbus/dbus-connection.c (_dbus_connection_init_id): initialize
749 the connection-global 33 bits of the object ID
751 * dbus/dbus-object-registry.c (info_from_entry): fill in
752 object ID in the new way
754 * dbus/dbus-objectid.h: rather than high/low bits, specifically
755 define server/client/instance bits.
757 2003-07-30 Havoc Pennington <hp@pobox.com>
759 * dbus/dbus-connection.c (dbus_connection_register_object): fix
762 2003-07-13 Havoc Pennington <hp@pobox.com>
764 * dbus/dbus-object.h (struct DBusObjectVTable): add padding
765 fields to DBusObjectVTable and DBusObjectInfo
767 2003-07-12 Havoc Pennington <hp@pobox.com>
769 * dbus/dbus-object-registry.c: implement unit test,
770 fix bugs discovered in process
772 * dbus/dbus-connection.c: remove handler_table and
773 register_handler(), add DBusObjectRegistry usage
775 * dbus/dbus-objectid.c (dbus_object_id_is_null)
776 (dbus_object_id_set_null): new functions
778 2003-07-08 Havoc Pennington <hp@pobox.com>
780 * dbus/dbus-object.c: implement some of this
782 * dbus/dbus-object-registry.c
783 (_dbus_object_registry_add_and_unlock): fill in the object_id out
785 (_dbus_object_registry_new): handle OOM
787 2003-07-08 Havoc Pennington <hp@pobox.com>
789 * dbus/dbus-object.h: sketch out an API for registering objects
790 with a connection, that allows us to use as little as 24 bytes
791 per object and lets application code represent an object in
794 * dbus/dbus-object-registry.c: implement the hard bits of the
795 DBusConnection aspect of object API. Not yet wired up.
797 2003-07-06 Havoc Pennington <hp@pobox.com>
799 * dbus/dbus-marshal.c (_dbus_marshal_set_object_id): new function
800 (_dbus_marshal_object_id): new
801 (_dbus_demarshal_object_id): new
802 (_dbus_marshal_get_arg_end_pos): support object ID type, and
803 consolidate identical switch cases. Don't conditionalize handling
804 of DBUS_TYPE_UINT64, need to handle the type always.
805 (_dbus_marshal_validate_arg): consolidate identical cases, and
806 handle DBUS_TYPE_OBJECT_ID
808 * dbus/dbus-objectid.c: new file with DBusObjectID data type.
810 * dbus/dbus-protocol.h: add DBUS_TYPE_OBJECT_ID
812 2003-09-28 Havoc Pennington <hp@pobox.com>
816 2003-09-28 Havoc Pennington <hp@pobox.com>
818 * doc/Makefile.am (dbus-specification.html): testing a funky hack
819 to work with Debian db2html
821 2003-09-28 Havoc Pennington <hp@pobox.com>
825 * doc/Makefile.am (dbus-test-plan.html): accept nonexistence of
826 stylesheet-images for benefit of Debian
828 Change back to using filesystem-linked sockets for the system
829 bus, so only root can create the default system bus address.
831 * bus/system.conf.in: change to use
832 DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
834 * dbus/Makefile.am (INCLUDES): remove DBUS_SYSTEM_BUS_PATH define
837 * configure.in: define DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
838 here, and AC_DEFINE DBUS_SYSTEM_PATH
840 2003-08-09 Anders Carlsson <andersca@codefactory.se>
846 2003-08-09 Anders Carlsson <andersca@codefactory.se>
848 * doc/dbus-specification.sgml:
849 Add activation reply values.
851 2003-08-05 Havoc Pennington <hp@redhat.com>
855 2003-08-05 Anders Carlsson <andersca@codefactory.se>
857 * glib/dbus-gmain.c: (watch_fd_new), (watch_fd_ref),
858 (watch_fd_unref), (dbus_gsource_check), (dbus_gsource_dispatch),
859 (add_watch), (remove_watch), (create_source):
860 Refcount fds, fixes some reentrancy issues.
862 2003-07-30 Havoc Pennington <hp@redhat.com>
864 * dbus/dbus-bus.c (init_connections_unlocked): fix default system
865 bus address to be abstract if we have abstract sockets
869 2003-07-28 Havoc Pennington <hp@redhat.com>
871 * bus/messagebus.in: fix to avoid processname/servicename
872 confusion, from Michael Kearey
873 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=100965
875 2003-07-23 Havoc Pennington <hp@pobox.com>
877 * dbus/dbus-message.c (dbus_message_iter_get_named):
878 fix from Andy Hanton to remove broken "+1"
880 2003-07-16 Havoc Pennington <hp@pobox.com>
882 * tools/dbus-launch.c (babysit): close stdout/stderr in the
883 babysitter process, as suggested by Thomas Leonard, so
884 an "eval `dbus-launch --exit-with-session`" will actually
887 2003-07-16 Havoc Pennington <hp@pobox.com>
889 * configure.in: print out EXPANDED_* variables in the summary at
890 the end; clean up the code that computes EXPANDED_ variables and
891 get the ones using exec_prefix right. Should make things work
892 when you build without --prefix
894 2003-06-29 Havoc Pennington <hp@pobox.com>
896 * mono/Test.cs (class Test): fire up a main loop and run it
898 * mono/DBus.cs (DBus): don't g_thread_init since it can only be
899 done once, the app has to do it
901 2003-06-26 Havoc Pennington <hp@pobox.com>
903 * mono/Connection.cs: set up connection with the glib main loop
905 2003-07-01 Havoc Pennington <hp@redhat.com>
907 * doc/dbus-specification.sgml: clarify the format of a type code,
908 change suggested by Jim Blandy
910 2003-06-29 Miloslav Trmac <mitr@volny.cz>
913 * tools/Makefile.am: Don't assume srcdir == builddir.
915 * dbus/dbus-memory.c (dbus_realloc): Don't check guards after shrinking
917 (_dbus_memory_test): New function.
918 * dbus/dbus-test.h: Add _dbus_memory_test ().
919 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): Call it.
921 * dbus/dbus-message.c (decode_header_data): Use %.4s instead
923 (dbus_message_new): Remove obsolete @todo.
925 * dbus/dbus-marshal.c (_dbus_marshal_set_int64)
926 (_dbus_marshal_set_uint64): Fix comment.
928 * dbus/dbus-message.c (append_int_field, append_uint_field): Don't
929 hardcode FIELD_REPLY_SERIAL.
931 * dbus/dbus-mainloop.c (_dbus_loop_remove_watch)
932 (_dbus_loop_remove_timeout): Cast function pointers to (void *) for %p
934 * configure.in: Add -D_POSIX_C_SOURCE=199309L -DBSD_SOURCE to CFLAGS
935 and disable DBUS_USE_ATOMIC_INT_486 when --enable-ansi is used
937 2003-06-24 Havoc Pennington <hp@pobox.com>
939 * mono/*.cs: Use IntPtr.Zero instead of ((IntPtr) 0)
941 2003-06-23 Anders Carlsson <andersca@codefactory.se>
947 * gcj/TestMessage.java: (TestMessage), (TestMessage.main):
948 * gcj/org/.cvsignore:
949 * gcj/org/Makefile.am:
950 * gcj/org/freedesktop/.cvsignore:
951 * gcj/org/freedesktop/Makefile.am:
952 * gcj/org/freedesktop/dbus/.cvsignore:
953 * gcj/org/freedesktop/dbus/Makefile.am:
954 * gcj/org/freedesktop/dbus/Message.java: (Message),
956 * gcj/org/freedesktop/dbus/natMessage.cc:
957 Fix the build system.
959 2003-06-22 Havoc Pennington <hp@pobox.com>
961 * mono/Connection.cs: add more bindings
963 * dbus/dbus-threads.c (dbus_threads_init): allow calling this
966 2003-06-22 Havoc Pennington <hp@pobox.com>
968 * mono/Connection.cs, mono/DBus.cs, mono/Error.cs:
969 Start wrapping more stuff.
971 2003-06-22 Havoc Pennington <hp@pobox.com>
973 * mono/Message.cs: implement Message.Wrap() that ensures we only
974 have a single C# wrapper per DBusMessage, assuming it works which
977 * dbus/dbus-message.c (dbus_message_allocate_data_slot): new
978 (dbus_message_free_data_slot): new
979 (dbus_message_set_data): new
980 (dbus_message_get_data): new
982 2003-06-22 Havoc Pennington <hp@pobox.com>
984 * dbus/dbus-dataslot.c (_dbus_data_slot_allocator_unref)
985 (_dbus_data_slot_allocator_alloc): rework these to keep a
986 reference count on each slot and automatically manage a global
987 slot ID variable passed in by address
989 * bus/bus.c: convert to new dataslot API
991 * dbus/dbus-bus.c: convert to new dataslot API
993 * dbus/dbus-connection.c: convert to new dataslot API
995 * dbus/dbus-server.c: convert to new dataslot API
997 * glib/dbus-gmain.c: ditto
1001 * bus/connection.c: ditto
1003 2003-06-22 Anders Carlsson <andersca@codefactory.se>
1005 * configure.in: Add AM_PROG_GCJ and move AM_PROG_LIBTOOL
1006 after the gcj checks so that the correct configuration tags
1007 will be added to libtool.
1009 * dbus-glib-1.pc.in: No need to specify any includes since
1010 dbus-1.pc.in has those.
1012 2003-06-22 Havoc Pennington <hp@pobox.com>
1014 * mono/*, gcj/*, configure.in, Makefile.am:
1015 Check in makefiles and subdirs for mono and gcj bindings.
1016 Neither binding actually exists, just trying to get through
1017 all the build and other boring bits.
1019 2003-06-21 Philip Blundell <philb@gnu.org>
1021 * tools/dbus-monitor.1: Updated.
1023 * tools/dbus-send.1: Likewise.
1025 2003-06-20 Anders Carlsson <andersca@codefactory.se>
1027 * dbus/dbus-transport-unix.c (unix_handle_watch): Check
1028 for hangup and error after checking read so we won't discard
1029 pending data if both hangup and read are set.
1031 2003-06-19 Philip Blundell <philb@gnu.org>
1033 * tools/dbus-print-message.c (print_message): Handle BOOLEAN.
1035 * tools/dbus-send.c: Accept both --system and --session.
1037 * tools/dbus-monitor.c: Same here.
1039 2003-06-19 Anders Carlsson <andersca@codefactory.se>
1041 * glib/dbus-glib.h: Fix so that dbus-glib.h can be used
1042 from C++ (Patch by Miloslav Trmac).
1044 2003-06-15 Joe Shaw <joe@assbarn.com>
1046 * configure.in: Check for socklen_t.
1048 * dbus/dbus-sysdeps.c: Define socklen_t if it's not defined.
1050 * test/test-segfault.c: Add #include <sys/time.h>
1052 * tools/Makefile.am: Add DBUS_X_CFLAGS to the INCLUDES since
1053 dbus-launch needs it.
1055 2003-06-09 Havoc Pennington <hp@redhat.com>
1057 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): don't use
1058 SUN_LEN, it breaks abstract socket usage
1060 * dbus/dbus-internals.c (_dbus_verbose_real): only print PID at
1063 2003-06-04 Havoc Pennington <hp@pobox.com>
1065 * dbus/dbus-server.c (dbus_server_listen): allow abstract sockets
1066 using unix:abstract=/foo, and when listening in a tmpdir
1067 i.e. unix:tmpdir=/tmp, always use abstract sockets if we can.
1069 * dbus/dbus-transport.c (_dbus_transport_open): support
1072 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
1073 support abstract sockets
1075 * dbus/dbus-transport-unix.c
1076 (_dbus_transport_new_for_domain_socket): support abstract sockets
1078 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket): add "abstract"
1079 toggle as an argument, implement abstract namespace support
1080 (_dbus_listen_unix_socket): ditto
1082 * configure.in: add --enable-abstract-sockets and implement
1083 a configure check for autodetection of the right value.
1085 2003-06-01 Havoc Pennington <hp@pobox.com>
1087 * tools/dbus-cleanup-sockets.c: add utility to clean up sockets
1088 in /tmp (though on Linux this will end up being useless,
1089 when we add abstract namespace support)
1091 * configure.in: define DBUS_SESSION_SOCKET_DIR in addition to
1094 2003-05-28 Colin Walters <walters@verbum.org>
1096 * tools/dbus-monitor.c (main): Fix silly typo (s/--session/--system/).
1098 2003-05-18 Anders Carlsson <andersca@codefactory.se>
1100 * dbus/dbus-message.c (dbus_message_new): Remove @todo.
1102 2003-05-17 Colin Walters <walters@gnu.org>
1104 * tools/dbus-send.c: Don't exit with an error code if --help was
1105 passed. Default to using the session bus instead of the system
1108 * tools/dbus-launch.c: Ditto.
1110 * tools/dbus-monitor.c: Ditto.
1112 * tools/dbus-send.1: Update with new arguments.
1114 * tools/dbus-launch.c: Emit code to export variables. New
1115 arguments -s and -c to specify shell syntax, and a bit of code to
1116 autodetect syntax. Also, allow specifying a program to run.
1118 * tools/dbus-launch.1: Update with new arguments.
1120 * tools/dbus-send.1: Ditto.
1122 * tools/dbus-monitor.1: Ditto.
1124 2003-05-17 Havoc Pennington <hp@pobox.com>
1126 * bus/config-parser.c (merge_included): merge in policies from
1127 child configuration file.
1129 * bus/policy.c (bus_policy_merge): function to merge two policies
1132 2003-05-16 Havoc Pennington <hp@redhat.com>
1134 * dbus/dbus-connection.c: disable verbose lock spew
1136 * tools/dbus-send.c: add --print-reply command line option
1138 * tools/dbus-print-message.h (print_message): new util function
1139 shared by dbus-send and dbus-monitor
1141 * tools/dbus-monitor.c (handler_func): exit on disconnect
1143 * dbus/dbus-transport-unix.c (do_reading): if the transport is
1144 disconnected, don't try to use the read_watch
1146 * dbus/dbus-watch.c (dbus_watch_get_enabled): assert watch != NULL
1147 so we can find this bug more easily
1149 2003-05-16 Havoc Pennington <hp@redhat.com>
1151 * bus/policy.c (free_rule_list_func): avoid a crash when passed
1152 NULL as DBusHashTable is annoyingly likely to do.
1154 2003-05-16 Colin Walters <walters@verbum.org>
1156 * tools/dbus-monitor.c: Add --session argument and usage()
1159 * tools/dbus-monitor.1: Update with new --session arg.
1161 * bus/Makefile.am (install-data-hook): Create
1162 $(libdir)/dbus-1.0/services so that the session bus is happy.
1164 2003-05-15 Havoc Pennington <hp@redhat.com>
1166 * dbus/dbus-sysdeps.c (_dbus_atomic_dec, _dbus_atomic_inc): work
1167 on non-x86. ifdef's are evil.
1169 2003-05-15 Havoc Pennington <hp@redhat.com>
1171 * configure.in: 0.11
1175 * bus/Makefile.am (initddir): apparently we are supposed to put
1176 init scripts in /etc/rc.d/init.d not /etc/init.d
1178 * bus/Makefile.am: remove the "you must --enable-tests to make
1179 check" as it broke distcheck
1181 * bus/Makefile.am (install-data-hook): create /etc/dbus-1/system.d
1183 2003-05-13 James Willcox <jwillcox@gnome.org>
1186 * bus/activation.c: (bus_activation_service_created),
1187 (bus_activation_activate_service):
1188 * bus/driver.c: (bus_driver_send_service_deleted),
1189 (bus_driver_send_service_created), (bus_driver_send_service_lost),
1190 (bus_driver_send_service_acquired),
1191 (bus_driver_send_welcome_message),
1192 (bus_driver_handle_list_services):
1193 * bus/session.conf.in:
1194 * dbus/dbus-bus.c: (dbus_bus_acquire_service),
1195 (dbus_bus_service_exists), (dbus_bus_activate_service):
1198 Add some convenience API which lets you activate a service, and did a
1199 bunch of s/0/DBUS_TYPE_INVALID/ in calls to dbus_message_append_args()
1200 and dbus_message_get_args()
1202 2003-05-11 Havoc Pennington <hp@pobox.com>
1204 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix to avoid
1205 calling _dbus_marshal_validate_arg() for every byte in a byte
1208 * dbus/dbus-message-handler.c: use atomic reference counting to
1209 reduce number of locks slightly; the global lock in here sucks
1211 * dbus/dbus-connection.c
1212 (_dbus_connection_update_dispatch_status_and_unlock): variant of
1213 update_dispatch_status that can be called with lock held; then use
1214 in a couple places to reduce locking/unlocking
1215 (dbus_connection_send): hold the lock over the whole function
1216 instead of acquiring it twice.
1218 * dbus/dbus-timeout.c (_dbus_timeout_new): handle OOM
1220 * bus/connection.c (bus_connections_setup_connection): fix access
1221 to already-freed memory.
1223 * dbus/dbus-connection.c: keep a little cache of linked list
1224 nodes, to avoid using the global linked list alloc lock in the
1225 normal send-message case. Instead we just use the connection lock
1226 that we already have to take.
1228 * dbus/dbus-list.c (_dbus_list_find_last): new function
1230 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec):
1231 change to use a struct for the atomic type; fix docs,
1232 they return value before increment, not after increment.
1234 * dbus/dbus-string.c (_dbus_string_append_4_aligned)
1235 (_dbus_string_append_8_aligned): new functions to try to
1236 microoptimize this operation.
1237 (reallocate_for_length): break this out of set_length(), to
1238 improve profile info, and also so we can consider inlining the
1241 * dbus/dbus-message.c (dbus_message_new_empty_header): init data
1242 strings with some preallocation, cuts down on our calls to realloc
1243 a fair bit. Though if we can get the "move entire string to empty
1244 string" optimization below to kick in here, it would be better.
1246 * dbus/dbus-string.c (_dbus_string_move): just call
1247 _dbus_string_move_len
1248 (_dbus_string_move_len): add a special case for moving
1249 an entire string into an empty string; we can just
1250 swap the string data instead of doing any reallocs.
1251 (_dbus_string_init_preallocated): new function
1253 2003-05-11 Havoc Pennington <hp@pobox.com>
1255 Write a "test-profile" that does echo client-server with threads;
1256 profile reveals lock contention, memcpy/realloc of buffers, and
1257 UTF-8 validation as hot spots. 20% of lock contention eliminated
1258 with dbus_atomic_inc/dec implementation on x86. Much remaining
1259 contention is global mempool locks for GList and DBusList.
1261 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec): add
1264 * dbus/dbus-connection.c (struct DBusConnection): use
1265 dbus_atomic_t for the reference count
1267 * dbus/dbus-message.c (struct DBusMessage): declare
1268 dbus_atomic_t values as volatile
1270 * configure.in: code to detect ability to use atomic integer
1271 operations in assembly, from GLib patch
1273 * dbus/dbus-internals.c (_dbus_verbose_real): call getpid every
1274 time, tired of it being wrong in threads and forked processes
1276 * glib/test-profile.c: a little program to bounce messages back
1277 and forth between threads and eat CPU
1279 * dbus/dbus-connection.c: add debug spew macros for debugging
1280 thread locks; include config.h at top; fix deadlock in
1281 dbus_connection_flush()
1283 2003-05-08 Havoc Pennington <hp@pobox.com>
1285 * dbus/dbus-spawn.c: s/_exit/exit/ because it was keeping gcov
1286 data from getting written, and there wasn't a good reason to
1289 * test/decode-gcov.c (mark_inside_dbus_build_tests): don't count
1290 dbus_verbose lines in test coverage
1291 (main): add list of functions sorted by # of untested blocks
1292 to the coverage report
1294 * dbus/dbus-mempool.c: put some test-only code in DBUS_BUILD_TESTS
1296 * dbus/dbus-marshal.c (_dbus_marshal_test): extend test coverage
1298 * dbus/dbus-message-handler.c (_dbus_message_handler_test):
1299 extend test coverage
1301 * test/data/auth/cancel.auth-script: test canceling an
1304 * dbus/Makefile.am: remove dbus-server-debug.[hc] for now, as they
1305 aren't used. in CVS history if we end up needing them.
1307 2003-05-04 Havoc Pennington <hp@pobox.com>
1309 * dbus/dbus-message-handler.c (_dbus_message_handler_test): add
1312 * dbus/dbus-marshal.c (_dbus_demarshal_string_array): fix this
1313 function, which assumed length was in # of strings, not bytes
1315 * dbus/dbus-message.c (_dbus_message_test): add tests for some
1318 * dbus/dbus-connection.c
1319 (_dbus_connection_queue_received_message): disable function for
1320 now, we are only using it in test mode
1322 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
1323 remove a mistaken FIXME
1325 2003-05-04 Havoc Pennington <hp@pobox.com>
1327 * dbus/dbus-connection.c (dbus_connection_preallocate_send):
1328 unlock mutex on successful return, patch from Anders Gustafsson
1330 2003-05-04 Havoc Pennington <hp@pobox.com>
1332 * dbus-glib-1.pc.in (Requires): fix dependencies, from
1335 2003-05-04 Havoc Pennington <hp@pobox.com>
1337 * tools/dbus-launch.c: implement
1339 * bus/main.c (main), bus/bus.c (bus_context_new):
1340 implement --print-pid and --fork
1342 2003-05-03 Havoc Pennington <hp@redhat.com>
1344 * dbus/dbus-address.c (dbus_parse_address): fix bug when a key in
1345 the address had no value, and add to test suite. Fix and
1346 regression test from Miloslav Trmac
1348 2003-05-03 Havoc Pennington <hp@pobox.com>
1350 * dbus/dbus-watch.c (dbus_watch_handle): warn and return if a
1351 watch is invalid when handled
1353 * tools/Makefile.am, tools/dbus-launch.c, tools/dbus-launch.1: add
1354 dbus-launch utility to launch the bus from a shell script. Didn't
1355 actually implement dbus-launch yet, it's just a placeholder still.
1357 2003-05-03 Havoc Pennington <hp@pobox.com>
1359 * bus/Makefile.am, bus/dbus-daemon-1.1.in: man page for the
1360 daemon; also documents daemon config file, so replaces
1361 doc/config-file.txt. Corrected some stuff from config-file.txt in
1362 the process of moving it.
1364 2003-05-03 Havoc Pennington <hp@pobox.com>
1366 * tools/Makefile.am, tools/dbus-send.1, tools/dbus-monitor.1:
1369 2003-05-03 Colin Walters <walters@verbum.org>
1371 * dbus/dbus-sysdeps.c (fill_user_info): Test against
1372 DBUS_UID_UNSET to determine whether to do a uid lookup or not.
1374 * Makefile.am: Update to use new .pc versioning scheme.
1376 2003-05-02 Havoc Pennington <hp@redhat.com>
1378 * bus/system.conf.in: allow send/receive to/from message bus
1381 2003-04-30 Havoc Pennington <hp@redhat.com>
1383 * configure.in: print a note when building with unit tests and
1386 2003-04-30 Havoc Pennington <hp@redhat.com>
1388 * Makefile.am: add a check-local that complains if you didn't
1389 configure with --enable-tests
1391 2003-04-29 Havoc Pennington <hp@redhat.com>
1393 * glib/dbus-gmain.c: docs cleanups
1395 * dbus/dbus-types.h: add docs on int64 types
1397 * dbus/dbus-memory.c: fix docs to avoid putting private API in
1398 public API docs section
1400 2003-04-29 Havoc Pennington <hp@redhat.com>
1402 * dbus-1.pc.in, dbus-glib-1.pc.in: rename these from
1403 dbus-1.0.pc.in, dbus-glib-1.0.pc.in. As these change with the
1404 parallel install API version, not with the D-BUS package version.
1406 * HACKING: move some of README over here
1408 * README: updates, and document API/ABI policy
1410 * configure.in: reindentation
1412 2003-04-29 Havoc Pennington <hp@redhat.com>
1414 * dbus/dbus.h: add "you have to define DBUS_API_SUBJECT_TO_CHANGE
1415 to use this library" to be sure people have the right
1418 2003-04-28 Havoc Pennington <hp@redhat.com>
1420 * configure.in: add --enable-docs which by default is auto yes if
1421 doxygen and db2html found, no otherwise; but can be forced on/off
1423 * doc/Makefile.am: conditionalize whether to build docs on
1426 2003-04-28 Havoc Pennington <hp@redhat.com>
1428 * configure.in: 0.10
1432 * bus/system.conf.in: add <includedir>system.d</includedir>
1434 * dbus/dbus-userdb.c (_dbus_user_database_lookup): fix bug when
1435 username was provided but not uid
1437 * bus/config-parser.c (struct BusConfigParser): keep track of
1438 whether the parser is toplevel or was included; change some
1439 of the error handling if it's included.
1441 2003-04-27 Havoc Pennington <hp@pobox.com>
1445 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
1446 report correct status if we finish processing authentication
1447 inside this function.
1449 * bus/activation.c (try_send_activation_failure): use
1450 bus_transaction_send_error_reply
1452 * bus/connection.c (bus_connection_get_groups): return an error
1453 explaining the problem
1455 * bus/bus.c (bus_context_check_security_policy): implement
1456 restriction here that inactive connections can only send the
1457 hello message. Also, allow bus driver to send anything to
1460 * bus/connection.c (bus_connection_complete): create the
1461 BusClientPolicy here instead of on-demand.
1462 (bus_connection_get_policy): don't return an error
1464 * dbus/dbus-message.c (dbus_message_new_error_reply): allow NULL
1465 sender field in message being replied to
1467 * bus/bus.c (bus_context_check_security_policy): fix silly typo
1468 causing it to return FALSE always
1470 * bus/policy.c (bus_client_policy_check_can_send): fix bug where
1471 we checked sender rather than destination
1473 2003-04-25 Havoc Pennington <hp@redhat.com>
1475 test suite is slightly hosed at the moment, will fix soon
1477 * bus/connection.c (bus_connections_expire_incomplete): fix to
1478 properly disable the timeout when required
1479 (bus_connection_set_name): check whether we can remove incomplete
1480 connections timeout after we complete each connection.
1482 * dbus/dbus-mainloop.c (check_timeout): fix this up a bit,
1483 probably still broken.
1485 * bus/services.c (bus_registry_acquire_service): implement max
1486 number of services owned, and honor allow/deny rules on which
1487 services a connection can own.
1489 * bus/connection.c (bus_connection_get_policy): report errors here
1491 * bus/activation.c: implement limit on number of pending
1494 2003-04-25 Havoc Pennington <hp@redhat.com>
1496 * dbus/dbus-transport.c (_dbus_transport_get_unix_user): fix bug
1497 where we used >= 0 instead of != DBUS_UID_UNSET.
1499 2003-04-25 Havoc Pennington <hp@redhat.com>
1501 * glib/dbus-gmain.c (remove_watch): fix for a crash when watches
1502 were toggled without add/remove, fix from Anders Gustafsson
1504 2003-04-24 Havoc Pennington <hp@redhat.com>
1506 * test/data/valid-config-files/basic.conf: add <limit> tags to
1509 * bus/config-parser.h, bus/config-parser.c, bus/bus.c: Implement
1510 <limit> tag in configuration file.
1512 2003-04-24 Havoc Pennington <hp@redhat.com>
1514 * bus/dispatch.c: somehow missed some name_is
1516 * dbus/dbus-timeout.c (_dbus_timeout_set_enabled)
1517 (_dbus_timeout_set_interval): new
1519 * bus/connection.c (bus_connections_setup_connection): record time
1520 when each connection is first set up, and expire them after the
1521 auth timeout passes.
1523 2003-04-24 Havoc Pennington <hp@redhat.com>
1525 * dbus/dbus-message.c (dbus_message_name_is): rename
1526 (dbus_message_service_is): rename
1527 (dbus_message_sender_is): rename
1528 (dbus_message_get_service): rename
1530 2003-04-24 Havoc Pennington <hp@redhat.com>
1532 * configure.in: add --enable-checks
1534 * dbus/dbus-message.c (dbus_message_new): reverse name/service arguments
1536 * dbus/dbus-connection.c (dbus_connection_preallocate_send): fix
1537 to use thread locks.
1538 (_dbus_connection_handler_destroyed_locked): move some private
1539 functions into proper docs group
1541 * dbus/dbus-internals.h: add _dbus_return_if_fail,
1542 _dbus_return_val_if_fail
1544 Throughout: use dbus_return_if_fail
1546 2003-04-23 James Willcox <jwillcox@gnome.org>
1549 * glib/dbus-gmain.c: (add_timeout), (wakeup_main), (create_source),
1550 (dbus_connection_setup_with_g_main),
1551 (dbus_server_setup_with_g_main):
1552 * glib/test-dbus-glib.c: (main):
1553 * glib/test-thread-client.c: (main):
1554 * glib/test-thread-server.c: (new_connection_callback), (main):
1555 * tools/dbus-monitor.c: (main):
1557 Added a GMainContext argument to dbus_connection_setup_with_g_main()
1558 and dbus_server_setup_with_g_main().
1560 2003-04-20 Havoc Pennington <hp@pobox.com>
1562 * doc/dbus-specification.sgml: document the restrictions on
1563 message and service names
1565 2003-04-22 Havoc Pennington <hp@redhat.com>
1567 * dbus/dbus-message.c, dbus/dbus-marshal.c: add 64-bit integer
1568 support, and do some code cleanups to share more code and
1569 speed up array marshal/demarshal.
1571 * dbus-1.0.pc.in (Cflags): put libdir include file in cflags
1573 * configure.in: generate dbus-arch-deps.h
1575 * dbus/dbus-protocol.h (DBUS_TYPE_INT64, DBUS_TYPE_UINT64): add
1578 2003-04-22 Havoc Pennington <hp@redhat.com>
1580 * test/data/valid-messages/opposite-endian.message: fix test
1581 to use proper type for rply field
1583 * test/data/invalid-messages: add tests for below validation
1585 * dbus/dbus-message.c (decode_header_data): validate field types,
1586 and validate that named fields are valid names
1587 (decode_name_field): consider messages in the
1588 org.freedesktop.Local. namespace to be invalid.
1590 * dbus/dbus-string.c (_dbus_string_validate_name): new
1592 2003-04-19 Havoc Pennington <hp@pobox.com>
1594 * bus/driver.c (bus_driver_handle_hello): check limits and
1595 return an error if they are exceeded.
1597 * bus/connection.c: maintain separate lists of active and inactive
1598 connections, and a count of each. Maintain count of completed
1599 connections per user. Implement code to check connection limits.
1601 * dbus/dbus-list.c (_dbus_list_unlink): export
1603 * bus/bus.c (bus_context_check_security_policy): enforce a maximum
1604 number of bytes in the message queue for a connection
1606 2003-04-18 Havoc Pennington <hp@pobox.com>
1608 * dbus/dbus-auth.c (record_mechanisms): memleak fixes
1610 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): fix some
1613 * dbus/dbus-keyring.c (add_new_key): fix a memleak, and
1614 on realloc be sure to update the pointer in the keyring
1616 * dbus/dbus-string.c (_dbus_string_zero): compensate for align
1617 offset to avoid writing to unallocated memory
1619 * dbus/dbus-auth.c (process_rejected): return FALSE if we fail to
1620 try the next mechanism, so we properly handle OOM
1622 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): fix double-free
1624 (_dbus_keyring_new): fix OOM bug
1625 (_dbus_keyring_new_homedir): always set error; impose a maximum
1626 number of keys we'll load from the file, mostly to speed up the
1627 test suite and make its OOM checks more useful, but also for
1630 * dbus/dbus-auth.c (process_error_server): reject authentication
1631 if we get an error from the client
1632 (process_cancel): on cancel, send REJECTED, per the spec
1633 (process_error_client): send CANCEL if we get an error from the
1636 2003-04-18 Havoc Pennington <hp@pobox.com>
1638 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix UMR in verbose
1641 * dbus/dbus-auth.c (handle_client_data_cookie_sha1_mech): fix OOM
1644 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): only whine
1645 about DBUS_TEST_HOMEDIR once
1647 * bus/Makefile.am (TESTS_ENVIRONMENT): put DBUS_TEST_HOMEDIR in
1650 * bus/dispatch.c (bus_dispatch_sha1_test): actually load sha1
1651 config file so we test the right thing
1653 Throughout: assorted docs improvements
1655 2003-04-18 Havoc Pennington <hp@pobox.com>
1657 * glib/dbus-gmain.c: adapt to watch changes
1659 * bus/bus.c, bus/activation.c, etc.: adjust to watch changes
1661 * dbus/dbus-server.h: remove dbus_server_handle_watch
1663 * dbus/dbus-connection.h: remove dbus_connection_handle_watch
1665 * dbus/dbus-watch.c (dbus_watch_handle): change DBusWatch to work
1666 like DBusTimeout, so we don't need dbus_connection_handle_watch
1669 2003-04-17 Havoc Pennington <hp@redhat.com>
1671 * dbus/dbus-userdb.c, dbus/dbus-sysdeps.c: redo all the passwd
1672 database usage so it all goes via the DBusUserDatabase cache.
1674 2003-04-17 Havoc Pennington <hp@redhat.com>
1676 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix logic so that if
1677 there was an OOM watch we skipped, we always return TRUE so we
1678 iterate again to have a look at it again. Fixes test suite hang.
1679 Code rearrangement also lets us lose some memset and only iterate
1680 over callbacks once.
1682 * bus/driver.c (bus_driver_handle_message): sense of test for
1685 2003-04-16 Havoc Pennington <hp@pobox.com>
1687 * doc/dbus-specification.sgml: make spec say serials are unsigned
1689 * dbus/dbus-message.h: change message serials to unsigned
1691 * dbus/dbus-connection.c: adapt to message serials being unsigned
1693 2003-04-15 Havoc Pennington <hp@pobox.com>
1695 * bus/bus.c: create and keep around a shared DBusUserDatabase
1698 * bus/connection.c (bus_connection_get_groups): don't cache
1699 groups for user in the connection object, since user database
1700 object now does that.
1702 2003-04-16 Havoc Pennington <hp@redhat.com>
1704 * dbus/dbus-message.c (_dbus_message_add_size_counter): keep a
1705 list of size counters
1706 (_dbus_message_loader_putback_message_link): put back a popped link
1708 * dbus/dbus-connection.c
1709 (dbus_connection_set_max_live_messages_size): rename
1711 (dbus_connection_get_outgoing_size): get size of outgoing
1713 (_dbus_connection_set_connection_counter): remove this cruft
1715 2003-04-14 Havoc Pennington <hp@redhat.com>
1717 * dbus/dbus-userdb.c: user database abstraction, mostly to get
1718 caching, but at some point we might want to be able to use a
1721 * bus/dispatch.c (bus_dispatch_sha1_test): add a test that uses
1722 SHA1 conf file to test the sha1 auth mechanism, since the regular
1723 test always uses EXTERNAL when available.
1726 test/data/valid-config-files/debug-allow-all-sha1.conf.in:
1727 add conf file that requires use of sha1 auth
1729 2003-04-13 Havoc Pennington <hp@pobox.com>
1731 * tools/dbus-send.c, tools/dbus-monitor.c: two utility programs
1732 from Philip Blundell to send messages and monitor them.
1734 2003-04-13 Havoc Pennington <hp@pobox.com>
1736 * dbus/dbus-mainloop.c: fix some reentrancy issues by refcounting
1739 * test/data/valid-config-files/debug-allow-all.conf.in: allow all
1742 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
1743 fix to only recover unused bytes if we're already authenticated
1744 (_dbus_transport_get_is_authenticated): fix to still mark us
1745 authenticated if there are unused bytes.
1747 * bus/dispatch.c: implement security policy checking
1749 * bus/connection.c (bus_transaction_send_from_driver): new
1751 * bus/bus.c (bus_context_check_security_policy): new
1753 * bus/dispatch.c (send_service_nonexistent_error): delete this,
1754 now we just set the DBusError and it gets converted to an error
1757 * bus/connection.c (allow_user_function): enable code using actual
1758 data from the config file
1760 * bus/policy.c (list_allows_user): handle wildcard rules for
1761 user/group connection perms
1763 2003-04-13 Havoc Pennington <hp@pobox.com>
1765 * bus/config-parser.c: Load up the BusPolicy and BusPolicyRules
1767 * dbus/dbus-sysdeps.c (_dbus_get_user_id): new function
1769 * bus/policy.c (bus_policy_append_mandatory_rule)
1770 (bus_policy_append_default_rule, bus_policy_append_user_rule)
1771 (bus_policy_append_group_rule): new functions
1773 2003-04-12 Havoc Pennington <hp@pobox.com>
1775 * bus/config-parser.c (bus_config_parser_new): fix a memleak
1777 * dbus/dbus-sysdeps.c: change DBusCredentials to use longs for
1778 the pid/gid/uid, just for paranoia.
1780 * test/break-loader.c (randomly_do_n_things): find a byte
1781 containing a type code, and randomly change it to a different
1784 2003-04-12 Havoc Pennington <hp@pobox.com>
1786 * bus/policy.h: change BusPolicy to be the thing from the config
1787 file, and rename old BusPolicy to BusClientPolicy
1789 * bus/bus.c, bus/connection.c, bus/config-parser.c: change to
1790 match change in how policy works
1792 * dbus/dbus-internals.h: mark assert_not_reached as
1793 __attribute((noreturn))__
1795 2003-04-11 Havoc Pennington <hp@redhat.com>
1797 * doc/dbus-specification.sgml: fix a spot with the wrong name for
1798 the broadcast service. Use boolean return for ServiceExists.
1800 2003-04-11 Havoc Pennington <hp@redhat.com>
1802 * configure.in: add another directory to look for qt in.
1804 2003-04-11 Havoc Pennington <hp@redhat.com>
1806 * AUTHORS: add Colin Walters
1808 2003-04-11 Havoc Pennington <hp@redhat.com>
1814 2003-04-11 Havoc Pennington <hp@redhat.com>
1816 * bus/messagebus.in: remove pid file when stopping the
1817 message bus, since the bus won't have privileges to remove it
1820 2003-04-11 Havoc Pennington <hp@redhat.com>
1822 * bus/bus.c (bus_context_new): move credentials change after
1825 2003-04-11 Havoc Pennington <hp@pobox.com>
1827 * test/decode-gcov.c: add "below average functions" to the
1828 coverage report, and change how some of the code works.
1830 * bus/test-main.c: bracket some stuff in DBUS_BUILD_TESTS so it's
1831 not in the coverage stats.
1833 * test/test-service.c (main): use _dbus_verbose not fprintf in a
1834 couple places so running the test suite doesn't result in megaspam.
1836 2003-04-11 Havoc Pennington <hp@pobox.com>
1838 * bus/dispatch.c (check_existent_service_activation): accept a no
1839 memory error in a place we didn't before
1841 * bus/test.c (bus_test_run_everything): remove hacky "do it twice
1842 in case the first one failed," since the test suite is less
1845 2003-04-10 Havoc Pennington <hp@pobox.com>
1847 * bus/dispatch.c (check_segfault_service_activation): add test
1848 for launching an executable that just crashes.
1850 * test/test-segfault.c (main): try setting coredumpsize to 0 so we
1851 don't leave a million cores. We'll see how portable this is.
1853 2003-04-10 Havoc Pennington <hp@pobox.com>
1855 * dbus/dbus-spawn.c (_dbus_spawn_async_with_babysitter): move all
1856 the possible parent failures before we fork, so that we don't
1857 fail to create a babysitter after creating the child.
1859 * bus/activation.c (bus_activation_activate_service): kill child
1860 if we don't successfully complete the activation.
1862 2003-04-10 Havoc Pennington <hp@redhat.com>
1864 * dbus/dbus-connection.c (dbus_connection_flush): don't spin on
1865 the connection if it's disconnected
1867 * bus/activation.c (bus_activation_service_created): use new
1868 transaction features to roll back removal of pending activation if
1869 we don't successfully create the service after all. Don't remove
1870 pending activation if the function fails.
1872 * dbus/dbus-list.c (_dbus_list_insert_before_link)
1873 (_dbus_list_insert_after_link): new code to facilitate
1876 * dbus/dbus-hash.c (_dbus_hash_table_insert_string_preallocated):
1877 new functionality, so we can preallocate the ability to insert
1880 * bus/connection.c (bus_transaction_add_cancel_hook): new function
1881 allowing us to put custom hooks in a transaction to be used for
1882 cancelling said transaction
1884 * doc/dbus-specification.sgml: add some discussion of secondary
1885 service owners, and disallow zero-length service names
1887 * bus/services.c (bus_registry_acquire_service): new function,
1888 splits out part of bus_driver_handle_acquire_service() and fixes
1889 a bug where we didn't remove the service doing the acquiring
1890 from the secondary queue if we failed to remove the current owner
1891 from the front of the queue.
1893 2003-04-10 Alexander Larsson <alexl@redhat.com>
1895 * doc/dbus-specification.sgml:
1896 s/org.freedesktop.Broadcast/org.freedesktop.DBus.Broadcast/
1898 2003-04-10 Alexander Larsson <alexl@redhat.com>
1903 Added files to cvsignore
1905 * dbus/dbus-message.h:
1906 * dbus/dbus-message.c: (dbus_message_iter_get_named):
1907 Make get_named() take two out argument and return a boolean.
1908 (dbus_message_iter_get_args_valist):
1909 Update usage of get_named().
1910 (dbus_message_iter_append_byte):
1912 (dbus_message_iter_append_named)
1914 (message_iter_test), (check_message_handling_type), (_dbus_message_test):
1917 2003-04-10 Alexander Larsson <alexl@redhat.com>
1919 * dbus/dbus-marshal.[ch]:
1920 Add array_type_pos argument to _dbus_marshal_validate_arg.
1921 Let you pass a NULL end_pos to _dbus_marshal_validate_type.
1923 * dbus/dbus-message.[ch]:
1924 Multi-dimensional arrays have full type specification in the
1925 outermost array. Iter code re-arranged to handle this.
1926 Added some more iter tests.
1928 * doc/dbus-specification.sgml:
1931 Update new array encoding description.
1932 Correct DBUS_SERVICE_FLAGS_REPLACE_EXISTING description.
1934 * test/data/invalid-messages/array-with-mixed-types.message:
1935 * test/data/valid-messages/array-of-array-of-uint32.message:
1936 Change to the new array format.
1938 * test/data/invalid-messages/too-short-dict.message:
1941 * test/data/valid-messages/recursive-types.message:
1942 Fix up and extend test.
1944 2003-04-10 Havoc Pennington <hp@pobox.com>
1946 * bus/dispatch.c: lots of fixes
1948 * dbus/dbus-mainloop.c (_dbus_loop_dispatch): export
1949 (_dbus_loop_iterate): remove old "quit if no callbacks" code,
1950 that was crack, broke the test service.
1952 * dbus/dbus-transport.c (_dbus_transport_open): fix error
1953 handling to avoid piling up errors if we get a failure on the
1956 * dbus/dbus-internals.c (_dbus_real_assert_not_reached): include
1957 pid in assertion failures.
1959 * dbus/dbus-mainloop.c (_dbus_loop_iterate): use static arrays up
1960 to some fixed size of file descriptor array. Don't return TRUE
1961 anytime a timeout exists, that led to lots of busy loop silliness
1964 2003-04-09 Havoc Pennington <hp@redhat.com>
1966 * dbus/dbus-mainloop.c (check_timeout): fix timeouts, I thought
1967 I'd checked this in earlier but hadn't.
1969 2003-04-09 Havoc Pennington <hp@redhat.com>
1971 * bus/dispatch.c (bus_dispatch_test): get a bit further through
1972 the activation test (man this is getting old!)
1974 2003-04-09 Havoc Pennington <hp@redhat.com>
1976 * test/test-utils.c: use dispatch status function to fix this up
1978 * bus/connection.c (connection_watch_callback): don't dispatch
1980 (connection_timeout_callback): don't dispatch from here
1981 (bus_connections_setup_connection): set the dispatch status function
1982 (bus_connection_disconnected): unset it
1984 * dbus/dbus-mainloop.c (_dbus_loop_queue_dispatch): new function
1985 used to add a connection to be dispatched
1986 (_dbus_loop_iterate): do the dispatching at the end of each
1989 * dbus/dbus-connection.c
1990 (dbus_connection_set_dispatch_status_function): new function
1991 allowing us to fix up main loop usage
1992 (_dbus_connection_last_unref): free all the various function
1994 (dbus_connection_dispatch): call the DispatchStatusFunction
1995 whenever this function returns
1996 (dbus_connection_handle_watch): call DispatchStatusFunction
1997 (dbus_connection_send_with_reply_and_block): call DispatchStatusFunction
1998 (reply_handler_timeout): call DispatchStatusFunction
1999 (dbus_connection_flush): call DispatchStatusFunction
2001 2003-04-09 Havoc Pennington <hp@redhat.com>
2003 * dbus/dbus-bus.c (dbus_bus_register): fix up error handling and
2006 * bus/dispatch.c (check_service_activated): fix bug in test
2008 * dbus/dbus-mainloop.c (check_timeout): fix this up
2010 * dbus/dbus-internals.c (_dbus_verbose_real): include PID in
2011 verbose output so we can sort out output from different processes,
2012 e.g. in the activation case.
2014 2003-04-08 Colin Walters <walters@gnu.org>
2016 * bus/bus.c (struct BusContext) [pidfile]: New member, to store
2018 (bus_context_new): Set it.
2019 (bus_context_unref): Use it to delete the pid file.
2021 2003-04-08 Havoc Pennington <hp@redhat.com>
2023 * test/data/invalid-messages/array-with-mixed-types.message:
2024 regression test that fails for the moment
2026 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): reorder
2027 tests for convenience
2029 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): don't allow
2030 array of nil, it broke things.
2032 * test/data/invalid-messages/array-of-nil.message: regression test
2034 * test/data/valid-messages/array-of-array-of-uint32.message:
2035 happened to write this so added it to suite
2037 2003-04-08 Havoc Pennington <hp@redhat.com>
2039 * bus/driver.c (bus_driver_handle_acquire_service): init
2040 retval/reply before checking name
2042 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add a
2043 recursion depth argument
2045 * dbus/dbus-message.h (struct DBusMessageIter): put some padding
2046 in the public struct for future extension
2048 * dbus/dbus-message-builder.c (_dbus_message_data_load): fix
2051 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix a verbose
2054 * doc/dbus-specification.sgml: fix typo
2056 2003-04-08 Alexander Larsson <alexl@redhat.com>
2058 Implemented recursive types, named types and new-style iters
2061 * glib/test-thread-client.c: (thread_func):
2062 * glib/test-thread-server.c: (handle_test_message):
2063 * test/test-service.c: (handle_echo):
2072 * dbus/dbus-internals.c: (_dbus_type_to_string):
2073 Update for new types.
2075 * dbus/dbus-marshal.[ch]:
2076 Implement recursive types and the new marshalling format.
2077 Remove hardcoded dict marshalling.
2078 Marshal named types.
2080 * dbus/dbus-message-builder.c:
2082 Remove references to old types
2084 * dbus/dbus-message.[ch]:
2085 New non-refcounted iter API that supports recursive iters.
2086 Use iters for appending, including support for recursive
2088 Add byte and named type support.
2089 Update everything to new marshalling formats.
2090 Add tests for new API.
2092 * dbus/dbus-protocol.h:
2093 Remove old array types.
2094 Add types: BYTE, ARRAY, DICT, NAMED
2096 * dbus/dbus-string.c:
2097 * dbus/dbus-sysdeps.c:
2098 Make parse_double locale safe.
2100 * dbus/dbus-test-main.c:
2106 * doc/dbus-specification.sgml:
2109 * test/data/incomplete-messages/missing-body.message:
2110 * test/data/invalid-messages/bad-boolean.message:
2111 * test/data/invalid-messages/bad-boolean-array.message:
2112 * test/data/invalid-messages/boolean-array-length-too-long.message-raw:
2113 * test/data/invalid-messages/boolean-has-no-value.message-raw:
2114 * test/data/invalid-messages/too-short-dict.message:
2115 * test/data/valid-messages/dict-simple.message:
2116 * test/data/valid-messages/dict.message:
2117 * test/data/valid-messages/emptiness.message:
2118 * test/data/valid-messages/lots-of-arguments.message:
2119 * test/data/valid-messages/no-padding.message:
2120 * test/data/valid-messages/recursive-types.message:
2121 Add missing NAME fields
2122 Fix up dicts & arrays
2124 * test/data/invalid-messages/dict-with-nil-value.message:
2125 Removed, this is not invalid anymore.
2127 * test/data/valid-messages/recursive-types.message:
2128 Add new test for deeply recursive types.
2130 2003-04-07 Havoc Pennington <hp@pobox.com>
2132 * bus/driver.c (bus_driver_handle_acquire_service): return an
2133 error if you try to acquire a service that starts with ':'
2135 2003-04-07 Havoc Pennington <hp@redhat.com>
2137 * doc/dbus-specification.sgml: require that base service names
2138 start with ':' and that the base service is created/deleted
2139 as first and last things a connection does on the bus
2141 * bus/dispatch.c (check_existent_service_activation): lots more
2142 work on the activation test; it doesn't fully pass yet...
2144 * test/test-service.c (main): fix so we don't memleak the
2145 connection to the message bus
2146 (filter_func): accept a message asking us to exit
2148 2003-04-06 Havoc Pennington <hp@pobox.com>
2150 * qt/Makefile.am (dbusinclude_HEADERS): install dbus-qt.h,
2153 * configure.in: fixes to Qt detection from Colin Walters
2155 * doc/Makefile.am: Only remove generated docbook dirs if they
2156 exist, from Colin Walters
2158 * dbus/dbus-bus.c: change how we set well-known connections to
2159 NULL, so that it works if a single connection is stored in
2160 two well-known array slots.
2162 * test/Makefile.am: remove a lot of stuff that isn't immediately
2163 useful, it's in CVS history if we want it.
2165 * test/test-service.c: use dbus-mainloop instead of that
2168 2003-04-06 Havoc Pennington <hp@pobox.com>
2170 * dbus/Makefile.am: split lists of sources into stuff that goes in
2171 the library, util functions that go in the lib and are also used
2172 elsewhere, and util functions that are used in tests/daemon but
2173 don't go in the lib.
2175 * dbus/dbus-mainloop.h, dbus/dbus-mainloop.c: move bus/loop.[hc]
2176 here so it can be used in test binaries also
2178 2003-04-06 Havoc Pennington <hp@pobox.com>
2180 * dbus/dbus-sysdeps.c (_dbus_become_daemon): write the pidfile
2181 here in the parent process, so we can return an error if it
2182 fails. Also, move some of the code into the child so the parent
2183 is less hosed if we fail midway through.
2185 * bus/bus.c (bus_context_new): move pidfile detection further up
2186 in the function, before we start overwriting sockets and such.
2188 * bus/messagebus.in: adjust this a bit, not sure if it will work.
2190 * configure.in: add --with-system-pid-file and --with-system-socket
2192 2003-04-06 Colin Walters <walters@verbum.org>
2194 * configure.in (DBUS_SYSTEM_PID_FILE): New variable.
2196 * bus/system.conf.in: Declare a pidfile.
2198 * bus/bus.c (bus_context_new): Test for an existing pid file, and
2199 create one (if appropriate).
2201 * bus/config-parser.c (enum ElementType) [ELEMENT_PIDFILE]: New.
2202 (struct BusConfigParser) [pidfile]: New.
2203 (element_type_to_name, merge_included, start_busconfig_child)
2204 (bus_config_parser_end_element, bus_config_parser_content): Handle it.
2205 (bus_config_parser_unref): Free it.
2206 (bus_config_parser_get_pidfile): New function.
2208 * bus/config-parser.h (_dbus_write_pid_file): Prototype.
2210 * dbus/dbus-errors.h (DBUS_ERROR_PIDFILE_EXISTS): New error.
2212 * dbus/dbus-sysdeps.c (_dbus_write_pid_file): New function.
2214 * dbus/dbus-sysdeps.h: Prototype it.
2216 2003-04-06 Havoc Pennington <hp@pobox.com>
2218 * bus/bus.c (bus_context_new): print the address in here, rather
2219 than in main(), because we need to do it before forking the daemon
2221 * bus/dispatch.c (send_service_nonexistent_error): set the sender
2222 on the service nonexistent error
2224 * bus/driver.c (bus_driver_handle_acquire_service): set the
2225 sender on the AcquireService reply
2227 * test/data/valid-config-files/debug-allow-all.conf.in: Make test
2228 server also listen on a UNIX socket so services can connect to it.
2230 2003-04-06 Havoc Pennington <hp@pobox.com>
2232 * dbus/dbus-threads.c: Redo how the fake debug mutexes are done
2233 so it detects deadlocks and also we actually init threads when
2236 2003-04-06 Havoc Pennington <hp@pobox.com>
2238 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
2239 save the domain socket name, and unlink it when we disconnect the
2240 server. Means that at least when we exit normally, we won't leave
2241 a bunch of junk in /tmp
2243 * dbus/dbus-transport-unix.c
2244 (_dbus_transport_new_for_domain_socket): code cleanup (nicer
2245 memory management). (I was making a real change here but then
2248 2003-04-06 Havoc Pennington <hp@pobox.com>
2250 * bus/bus.c (bus_context_new): fix wrong handling of
2251 server_data_slot_unref() in the error case.
2253 * dbus/dbus-internals.h (_dbus_assert): change so it passes
2254 "(condition) != 0" to _dbus_real_assert so that
2255 "_dbus_assert (pointer)" doesn't cause a warning
2257 * bus/main.c (main): accept --print-address option to print out
2258 the message bus address
2260 * dbus/dbus-sysdeps.c (_dbus_generate_random_ascii): export this
2262 * dbus/dbus-transport.c (_dbus_transport_open): special error for
2263 "tmpdir" option to unix: address on client side
2265 * dbus/dbus-server.c (dbus_server_listen): handle "tmpdir" option
2268 * configure.in (TEST_SOCKET_DIR): locate a temporary directory
2269 we can use to create sockets in the test suite.
2271 * bus/main.c (signal_handler): on SIGTERM, exit the daemon
2272 cleanly. To be used for testing.
2274 * dbus/dbus-spawn.c (babysit): use _dbus_set_signal_handler()
2276 * dbus/dbus-sysdeps.c (_dbus_set_signal_handler): new
2278 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
2279 handle trying to call this when there's no servers active
2281 2003-04-05 Havoc Pennington <hp@pobox.com>
2287 2003-04-05 Havoc Pennington <hp@pobox.com>
2289 * bus/bus.c (setup_server): fix this so dbus-daemon-1 doesn't
2290 crash on startup. Need to get "try starting the daemon"
2291 in the test suite I guess. ;-)
2293 * dbus/dbus-server.h, dbus/dbus-server.c: remove the stuff that
2294 tracked the number of open connections; it's better done in
2295 application-specific code as you want it to span all servers etc.
2297 2003-04-05 Havoc Pennington <hp@pobox.com>
2299 * bus/Makefile.am (install-data-hook): add missing DESTDIR,
2300 patch from Colin Walters
2302 2003-04-05 Havoc Pennington <hp@pobox.com>
2304 * doc/config-file.txt (Elements): fix docs of <auth> to reflect
2305 reality; in fact multiple mechanisms are allowed.
2307 * dbus/dbus-internals.c (_dbus_real_assert)
2308 (_dbus_real_assert_not_reached): move guts of _dbus_assert() and
2309 _dbus_assert_not_reached() into functions, so that they don't show
2310 up in basic block counts for test coverage, and don't use up as
2311 much disk space. Does mean slower execution speed though, so
2312 assumes --disable-asserts is the normal production case.
2314 2003-04-05 Havoc Pennington <hp@pobox.com>
2316 * test/Makefile.am (dist-hook): also dist *.in files
2322 2003-04-05 Havoc Pennington <hp@pobox.com>
2324 * dbus/dbus-string.c: docs warning
2326 * dbus/dbus-spawn.c: missing docs
2328 * dbus/dbus-memory.c (struct ShutdownClosure): missing docs
2330 2003-04-05 Havoc Pennington <hp@pobox.com>
2332 * bus/loop.c (bus_loop_iterate): fix the timeout code, using
2335 * dbus/dbus-spawn.c (_dbus_babysitter_unref): set sitter_pid
2336 to -1 once we've reaped the babysitter
2337 (_dbus_babysitter_handle_watch): do as much work as we can, not
2340 * bus/activation.c: add code using DBusBabysitter so that we
2341 handle it when a service fails to start up properly.
2342 (bus_activation_service_created): don't remove the activation
2343 entries as we go, just let them get removed when we free the pending
2344 activation. Unref reply messages after sending them.
2346 2003-04-05 Havoc Pennington <hp@pobox.com>
2348 * test/decode-gcov.c (main): print per-directory stats in the report
2350 * Makefile.am (coverage-report.txt): don't include test/* in gcov stats
2352 2003-04-05 Havoc Pennington <hp@pobox.com>
2354 * Makefile.am (coverage-report.txt): add target "coverage-report.txt"
2356 * test/decode-gcov.c: hack up a little program to suck data
2357 out of gcov files. Yes this is sort of silly.
2359 * configure.in: define something in config.h and do an
2360 AM_CONDITIONAL when gcov is enabled
2362 2003-04-04 Havoc Pennington <hp@redhat.com>
2364 * dbus/dbus-spawn.c, dbus/dbus-spawn.h: Change dbus_spawn to
2365 return a "babysitter" object that is used to monitor the status of
2366 the spawned process and reap it when required.
2368 * test/test-segfault.c, test/test-exit.c,
2369 test/test-sleep-forever.c: binaries that do various lame things,
2370 used in the test suite.
2372 * dbus/dbus-sysdeps.c: kill _dbus_errno_to_string()
2374 2003-04-03 Havoc Pennington <hp@pobox.com>
2376 * dbus/dbus-spawn.c: Move dbus-spawn into a separate file
2377 in preparation for modifying it, dbus-sysdeps is getting
2380 2003-04-03 Havoc Pennington <hp@redhat.com>
2382 * bus/loop.h, bus/loop.c: make the mainloop an object so we can
2385 * bus/*.[hc]: adapt to mainloop change
2387 2003-04-03 Havoc Pennington <hp@redhat.com>
2389 * bus/activation.c (load_directory): fix up memleaks
2390 (bus_activation_entry_free): free the entry
2392 * dbus/dbus-bus.c (dbus_bus_acquire_service): return an error if
2393 we get one from the message bus; fix memleaks.
2395 * dbus/dbus-message.c (dbus_set_error_from_message): new function
2397 2003-04-03 Havoc Pennington <hp@pobox.com>
2399 * bus/config-parser.c (bus_config_parser_unref): free
2400 list of mechanisms, bug discovered by test suite enhancements
2401 (putting system.conf and session.conf into suite)
2403 * test/Makefile.am, test/test-service.c: add placeholder for a
2404 test service that we'll activate as part of test suite. Doesn't
2407 * dbus/dbus-sysdeps.c (_dbus_setenv): support unsetenv by
2408 setting NULL value, and use system malloc not dbus_malloc()
2409 when we have unavoidable memleakage.
2411 * dbus/dbus-bus.c (dbus_bus_get): fix bug where bus type of 0
2412 didn't work, and support DBUS_BUS_ACTIVATION.
2414 * bus/activation.c (child_setup): pass our well-known bus type to
2417 * bus/config-parser.c: support <type> to specify well-known type
2419 * doc/dbus-specification.sgml: document the env variables to
2420 locate well-known buses and find service activator
2422 2003-04-02 Havoc Pennington <hp@redhat.com>
2424 * test/Makefile.am (all-local): add a rule to copy tests to
2425 builddir, so we can have generated tests. Use this to remove the
2426 silly hack for testing system.conf and session.conf. Will use this
2427 shortly to generate .service files pointing to test binaries.
2429 2003-04-02 Havoc Pennington <hp@redhat.com>
2431 * dbus/dbus-string.c (set_length): fix a bug - we allocated max of
2432 current alloc and needed new length, not max of the doubled
2433 allocation and needed new length. Also, when building tests,
2434 don't do the double-allocation stuff, just realloc every time.
2436 2003-04-02 Havoc Pennington <hp@redhat.com>
2438 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): include filenames
2440 (_dbus_string_get_dirname): new
2441 (_dbus_sysdeps_test): new
2442 (_dbus_directory_open): include dirnames in error messages
2444 * bus/config-parser.c: interpret <include> and <includedir> and
2445 <servicedir> relative to config file location if the given
2446 filename is not absolute.
2448 * dbus/dbus-string.c (_dbus_string_find_byte_backward): new
2450 2003-04-02 Havoc Pennington <hp@redhat.com>
2452 * bus/connection.c (bus_transaction_send_error_reply): set sender
2453 service for the error, and unref the reply on success
2455 * bus/activation.c: convert to use BusTransaction so OOM can be
2457 (bus_activation_service_created): set sender of the message
2459 2003-04-01 Havoc Pennington <hp@redhat.com>
2461 * bus/config-parser.c, bus/bus.c: implement <servicedir> and
2462 <includedir> (at least mostly)
2464 * dbus/dbus-sysdeps.c (_dbus_change_identity): set the group ID
2465 first, then the user ID
2467 2003-04-01 Havoc Pennington <hp@pobox.com>
2469 * dbus/dbus-server.c (dbus_server_set_auth_mechanisms): new
2472 * dbus/dbus-auth.c (_dbus_auth_set_mechanisms): new
2474 * dbus/dbus-internals.c (_dbus_dup_string_array): new function
2476 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): chmod the
2477 socket 0777, and unlink any existing socket.
2479 * bus/bus.c (bus_context_new): change our UID/GID and fork if
2480 the configuration file so specifies; set up auth mechanism
2483 * bus/config-parser.c (bus_config_parser_content): add support
2484 for <fork> option and fill in code for <auth>
2486 * bus/system.conf.in: add <fork/> to default configuration,
2487 and limit auth mechanisms to EXTERNAL
2489 * doc/config-file.txt (Elements): add <fork>
2491 * dbus/dbus-sysdeps.c (_dbus_become_daemon): new function
2492 (_dbus_change_identity): new function
2494 2003-03-31 Havoc Pennington <hp@redhat.com>
2496 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket)
2497 (_dbus_listen_unix_socket): fix off-by-one error in null
2498 termination spotted by Nalin
2500 2003-03-31 Havoc Pennington <hp@redhat.com>
2502 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): allow setting
2503 DBUS_TEST_HOMEDIR when tests are enabled, so we can test without
2504 having a real home directory available.
2506 2003-03-31 Havoc Pennington <hp@redhat.com>
2508 * bus/Makefile.am (install-data-hook): create /var/run/dbus
2510 * bus/messagebus.in: add init script for Red Hat /etc/init.d
2512 * configure.in: add support for specifying a style of init script
2515 2003-03-31 Havoc Pennington <hp@redhat.com>
2517 Fix some annoying DBusString API and fix all affected code.
2519 * dbus/dbus-string.c (_dbus_string_init): get rid of annoying
2521 (_dbus_string_get_data): change to return string instead of using
2523 (_dbus_string_get_const_data): ditto
2524 (_dbus_string_get_data_len): ditto
2525 (_dbus_string_get_const_data_len): ditto
2527 2003-03-31 Havoc Pennington <hp@redhat.com>
2529 * bus/main.c (main): fix up the command line arguments to be nicer
2531 2003-03-31 Havoc Pennington <hp@redhat.com>
2533 * dbus/Makefile.am (INCLUDES): use EXPANDED_LOCALSTATEDIR to
2534 define DBUS_SYSTEM_BUS_PATH as we want to compile in the same
2535 final location that lands in the config file
2537 * bus/config-loader-expat.c (bus_config_load): fix type of
2540 * doc/TODO: remove TODO item for dbus_bus_get()
2542 * dbus/dbus-bus.c (bus_data_free): add missing lock/unlock
2544 2003-03-31 Havoc Pennington <hp@pobox.com>
2546 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_domain_socket)
2547 (_dbus_transport_new_for_tcp_socket): these didn't need the "server"
2548 argument since they are always client side
2550 * dbus/dbus-server.c (dbus_server_get_address): new function
2552 * bus/main.c (main): take the configuration file as an argument.
2554 * test/data/valid-config-files/debug-allow-all.conf: new file to
2555 use with dispatch.c tests for example
2557 * bus/test-main.c (main): require test data dir
2559 * bus/bus.c (bus_context_new): change this to take a
2560 configuration file name as argument
2562 * doc/config-file.txt (Elements): add <servicedir>
2564 * bus/system.conf, bus/session.conf: new files
2566 * dbus/dbus-bus.c (dbus_bus_get): look for system bus on
2567 well-known socket if none set
2569 * configure.in: create system.conf and session.conf
2571 2003-03-30 Havoc Pennington <hp@pobox.com>
2573 * bus/config-parser.c: hacking
2575 * dbus/dbus-memory.c: don't use DBusList for the list of stuff
2576 to shut down, since it could cause weirdness with the DBusList
2579 * dbus/dbus-list.c (_dbus_list_test): add tests for the
2580 link-oriented stack routines
2581 (alloc_link): free the mempool if the first alloc from it fails
2583 * dbus/dbus-mempool.c (struct DBusMemBlock): fix alignment issue
2585 * dbus/dbus-string.c (UNICODE_VALID): sync new version of this
2587 (_dbus_string_skip_white): new
2589 * doc/config-file.txt (Elements): add <includedir>
2591 2003-03-28 Havoc Pennington <hp@pobox.com>
2593 * dbus/dbus-string.c (_dbus_string_copy_data_len)
2594 (_dbus_string_copy_data): new functions
2596 2003-03-28 Anders Carlsson <andersca@codefactory.se>
2598 * dbus/dbus-bus.c: (bus_data_free), (dbus_bus_get):
2602 * dbus/dbus-memory.c:
2605 2003-03-28 Havoc Pennington <hp@pobox.com>
2607 * bus/test.c (bus_test_flush_bus): remove the sleep from here,
2608 I think it may have just been superstition. Not sure.
2610 * dbus/dbus-string.c (_dbus_string_base64_decode): catch some OOM
2611 failures that were not being handled.
2613 * dbus/dbus-auth.c (process_auth): fix a memleak in OOM handling
2615 * dbus/dbus-memory.c: add ability to set number of mallocs in a
2616 row that will fail on out-of-memory.
2618 * dbus/dbus-internals.c (_dbus_test_oom_handling): convenience
2619 function for testing out-of-memory handling.
2621 * bus/config-loader-expat.c (memsuite): don't wrap the dbus
2622 allocation functions, they do map exactly to the expat ones.
2624 2003-03-27 Havoc Pennington <hp@redhat.com>
2626 * bus/config-loader-libxml.c (bus_config_load): add another error
2629 2003-03-26 Anders Carlsson <andersca@codefactory.se>
2632 Add note about automatic service activation.
2634 * doc/dbus-specification.sgml:
2635 Rename the specification and clarify a few things.
2637 2003-03-26 Anders Carlsson <andersca@codefactory.se>
2640 * dbus/dbus-address.c:
2642 * dbus/dbus-marshal.c:
2643 * dbus/dbus-server-debug-pipe.c:
2644 * dbus/dbus-transport-unix.c:
2645 Fix documentation warnings.
2647 2003-03-26 Havoc Pennington <hp@pobox.com>
2649 * bus/test-main.c, dbus/dbus-test.c (main): check memleaks
2650 after every test so it's quick and easy to see which leaked, and
2651 so we test multiple dbus_shutdown() calls
2653 * configure.in: change configure.in XML stuff to also support
2656 * config-loader-libxml.c: some hacking
2658 * config-loader-expat.c: some hacking
2660 * config-parser.c: some hacking, plus tests
2662 2003-03-25 Havoc Pennington <hp@redhat.com>
2664 * throughout - add more _DBUS_ASSERT_ERROR_IS_CLEAR
2666 * configure.in: add --with-xml option to specify XML library,
2667 right now only libxml is supported.
2669 * bus/config-loader-libxml.c, config-parser.c: sync some minor
2670 nonworking code between home and work, still just stubs
2672 2003-03-24 Havoc Pennington <hp@redhat.com>
2674 * dbus/dbus-sysdeps.c (_dbus_set_fd_nonblocking): move to this
2677 * dbus/dbus-errors.c (dbus_set_error, dbus_set_error_const): allow
2678 NULL argument for "message" if the error is a well-known one,
2679 fill in a generic message in this case.
2681 * dbus/dbus-errors.h (DBusResultCode): Kill DBusResultCode in
2684 * bus/test.c (bus_test_flush_bus): add
2686 * bus/policy.c (bus_policy_test): test code stub
2688 2003-03-24 Havoc Pennington <hp@pobox.com>
2690 * bus/connection.c (bus_connections_setup_connection): set up
2691 the "can this user connect" function, but it always returns
2692 TRUE until we have a config file parser so we can have a config
2693 file that allows connections.
2695 2003-03-23 Havoc Pennington <hp@pobox.com>
2697 * dbus/dbus-threads.c (dbus_mutex_new, dbus_condvar_new): with
2698 DBUS_BUILD_TESTS, actually alloc/free a block of memory for
2699 the mutex, so we can check for proper memory management
2702 * dbus/dbus-dataslot.c: remove the mutex from
2703 DBusDataSlotAllocator and lock it manually when using it,
2704 to simplify fitting it into the global slots framework.
2706 * dbus/dbus-threads.c (init_static_locks): rework how we're
2707 handling global locks so they are easily shut down.
2709 * bus/policy.c (bus_policy_append_rule): fix
2711 * bus/test-main.c (main): check for memleaks
2713 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): make
2714 test suite check for memleaks
2716 * dbus/dbus-memory.c: add support in test mode for tracking
2717 number of outstanding blocks
2719 2003-03-23 Havoc Pennington <hp@pobox.com>
2721 * bus/policy.c, bus/bus.c, bus/connection.c: implement allow/deny
2724 * dbus/dbus-hash.h: add ULONG hash keys
2726 * dbus/dbus-sysdeps.c (_dbus_get_groups): new
2727 (_dbus_get_group_id): new function
2729 2003-03-20 Havoc Pennington <hp@redhat.com>
2731 * dbus/dbus-connection.c (dbus_connection_set_unix_user_function):
2733 (dbus_connection_get_unix_user): new function
2735 2003-03-20 Havoc Pennington <hp@pobox.com>
2737 * bus/connection.c (bus_connection_send_oom_error): assert that
2738 message has a sender
2739 (connection_execute_transaction): ditto
2740 (bus_connection_preallocate_oom_error): fix to set the sender, and
2741 set recipient to the destination service, not the bus driver
2743 * bus/policy.c: hacking
2745 * dbus/dbus-message.c (dbus_message_service_is): new function
2746 (dbus_message_sender_is): new
2748 2003-03-19 Havoc Pennington <hp@redhat.com>
2750 * bus/policy.c: start sketching code for policy restrictions on
2751 what connections can do.
2753 2003-03-18 Havoc Pennington <hp@redhat.com>
2755 * doc/TODO: some notes on high-level todo items. Little nitpick
2756 stuff is all in @todo, so no need to add it here.
2758 * doc/config-file.txt: some notes on how config file might look
2760 2003-03-18 Anders Carlsson <andersca@codefactory.se>
2766 2003-03-17 Havoc Pennington <hp@redhat.com>
2768 * dbus/dbus-internals.h: add gcc attributes so that
2769 our printf-style functions warn on bad arguments to
2772 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): fix printf
2775 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
2778 2003-03-17 Havoc Pennington <hp@redhat.com>
2780 * bus/test-main.c (main): make it print something as it runs
2781 so make check doesn't look stuck
2783 * doc/negotiation.txt, doc/dbus-sasl-profile.txt: remove
2784 from CVS, now obsolete
2786 2003-03-17 Anders Carlsson <andersca@codefactory.se>
2788 * bus/dispatch.c: (bus_dispatch):
2789 Refetch the service name since it may have been reallocated
2790 when dbus_message_set_sender was called.
2792 * dbus/dbus-sysdeps.c: (_dbus_accept):
2793 Add address and address length variables and use them to stop
2794 valgrind from complaining.
2796 2003-03-17 Havoc Pennington <hp@pobox.com>
2798 All tests pass, no memleaks, no valgrind complaints.
2800 * bus/test.c: refcount handler_slot
2802 * bus/connection.c (bus_connections_new): refcount
2803 connection_data_slot
2805 * dbus/dbus-auth-script.c (_dbus_auth_script_run): delete unused
2806 bytes so that auth scripts pass.
2808 * bus/dispatch.c: init message_handler_slot so it gets allocated
2811 * bus/dispatch.c (message_handler_slot_ref): fix memleak
2813 * dbus/dbus-server-debug-pipe.c (_dbus_server_debug_pipe_new):
2814 dealloc server_pipe_hash when no longer used for benefit of
2817 * dbus/dbus-auth.c (process_command): memleak fix
2819 * bus/dispatch.c (check_hello_message): memleak fix
2821 2003-03-16 Havoc Pennington <hp@pobox.com>
2823 * dbus/dbus-bus.c (ensure_bus_data): fix double-unref of the data slot
2825 2003-03-17 Anders Carlsson <andersca@codefactory.se>
2827 * bus/activation.c (bus_activation_activate_service): Append
2828 the pending activation entry to the list of pending activations.
2830 2003-03-16 Havoc Pennington <hp@pobox.com>
2832 * bus/dispatch.c (bus_dispatch_test): remove double-unrefs of
2835 * dbus/dbus-address.c (create_entry): fix OOM handling when
2836 failing to alloc entry->method
2838 2003-03-16 Havoc Pennington <hp@pobox.com>
2840 * dbus/dbus-watch.c (_dbus_watch_new): handle failure to malloc
2843 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
2844 add some missing dbus_set_result
2846 * bus/dispatch.c (bus_dispatch_add_connection): handle failure to
2847 alloc the DBusMessageHandler
2849 * dbus/dbus-transport.c (_dbus_transport_disconnect): don't ref
2850 the transport here, since we call this from the finalizer; it
2851 resulted in a double-finalize.
2853 * dbus/dbus-transport.c (_dbus_transport_disconnect): fix a bug
2854 where we tried to use transport->connection that was NULL,
2855 happened when transport was disconnected early on due to OOM
2857 * bus/*.c: adapt to handle OOM for watches/timeouts
2859 * dbus/dbus-transport-unix.c: port to handle OOM during
2862 * dbus/dbus-auth.c (_dbus_auth_get_unused_bytes): return a
2863 reference to unused bytes instead of a copy
2865 * dbus/dbus-server.c (dbus_server_handle_watch): return FALSE for
2868 * dbus/dbus-connection.c (dbus_connection_handle_watch): return
2871 * dbus/dbus-timeout.c (dbus_timeout_handle): return FALSE for out
2874 2003-03-16 Anders Carlsson <andersca@codefactory.se>
2876 * doc/dbus-specification.sgml:
2877 Document reply message for ActivateService.
2879 2003-03-16 Anders Carlsson <andersca@codefactory.se>
2881 * bus/activation.c: (bus_pending_activation_entry_free),
2882 (bus_pending_activation_free), (bus_activation_new),
2883 (bus_activation_unref), (bus_activation_service_created),
2884 (bus_activation_activate_service):
2886 * bus/bus.c: (bus_context_new):
2887 * bus/desktop-file.c: (new_section):
2888 * bus/driver.c: (bus_driver_send_service_deleted),
2889 (bus_driver_handle_activate_service):
2890 * bus/services.c: (bus_registry_new), (bus_registry_ensure):
2892 * dbus/dbus-connection.c:
2893 (dbus_connection_send_with_reply_and_block):
2894 * dbus/dbus-message.c: (dbus_message_append_args_valist):
2895 * dbus/dbus-protocol.h:
2896 Make activation work better. Now pending activations will be queued
2897 and the daemon won't try to activate services that are already registered.
2899 2003-03-16 Havoc Pennington <hp@pobox.com>
2901 * dbus/dbus-bus.c (ensure_bus_data): handle failure to set
2904 * dbus/dbus-memory.c (_dbus_initialize_malloc_debug): support
2905 DBUS_MALLOC_BACKTRACES to print trace when failing an alloc
2907 2003-03-16 Havoc Pennington <hp@pobox.com>
2909 * dbus/dbus-string.c (_dbus_string_validate_utf8): oops, unbreak
2910 this. always run the test suite before commit...
2912 * bus/*: adapt to DBusConnection API changes
2914 * glib/dbus-gmain.c: adapt to DBusConnection API changes,
2915 requires renaming stuff to avoid dbus_connection_dispatch name
2918 * dbus/dbus-transport.c (_dbus_transport_queue_messages): new
2921 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
2922 separate from _dbus_message_loader_return_buffer()
2924 * dbus/dbus-connection.c (dbus_connection_get_n_messages): remove
2925 this, because it's now always broken to use; the number of
2926 messages in queue vs. the number still buffered by the message
2927 loader is undefined/meaningless. Should use
2928 dbus_connection_get_dispatch_state().
2929 (dbus_connection_dispatch): rename from
2930 dbus_connection_dispatch_message
2932 2003-03-16 Havoc Pennington <hp@pobox.com>
2934 * dbus/dbus-string.c (_dbus_string_validate_utf8): copy in a real
2937 2003-03-16 Anders Carlsson <andersca@codefactory.se>
2939 * dbus/dbus-connection.c:
2940 (dbus_connection_send_with_reply_and_block):
2941 Decrease connection->n_incoming when removing an entry
2943 * dbus/dbus-dict.c: (dbus_dict_entry_free),
2944 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
2945 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
2946 (dbus_dict_set_byte_array), (dbus_dict_set_string_array),
2947 (dbus_dict_get_boolean_array), (dbus_dict_get_double_array),
2948 (dbus_dict_get_byte_array):
2949 Handle NULL arrays and strings. Also add support for byte arrays.
2951 * dbus/dbus-marshal.c: (_dbus_marshal_byte_array),
2952 (_dbus_marshal_dict), (_dbus_demarshal_byte_array),
2953 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
2954 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
2955 (_dbus_demarshal_dict), (demarshal_and_validate_len),
2956 (_dbus_marshal_validate_arg), (_dbus_marshal_test):
2957 * dbus/dbus-marshal.h:
2958 Add support for marshalling and demarshalling empty arrays and strings.
2960 * dbus/dbus-message.c: (dbus_message_append_args_valist),
2961 (dbus_message_append_string_array),
2962 (dbus_message_iter_get_boolean),
2963 (dbus_message_iter_get_boolean_array),
2964 (dbus_message_iter_get_int32_array),
2965 (dbus_message_iter_get_uint32_array),
2966 (dbus_message_iter_get_double_array),
2967 (dbus_message_iter_get_byte_array),
2968 (dbus_message_iter_get_string_array), (dbus_message_iter_get_dict),
2969 (check_message_handling):
2970 Add support for getting empty arrays and dicts.
2972 * dbus/dbus-string.c: (_dbus_string_validate_utf8):
2973 Don't do any validation at all for now, that's better than just checking
2976 * test/data/valid-messages/emptiness.message:
2977 New test message with lots of empty arrays.
2979 2003-03-16 Havoc Pennington <hp@pobox.com>
2981 * dbus/dbus-connection.c
2982 (_dbus_connection_queue_received_message_link): new function that
2983 can't fail due to OOM
2985 * dbus/dbus-message.c (_dbus_message_loader_pop_message_link):
2986 new function pops a message together with a list link
2989 * dbus/dbus-transport-unix.c (queue_messages): use new link-based
2990 message queuing functions to avoid needing to alloc memory
2992 2003-03-16 Havoc Pennington <hp@pobox.com>
2994 Oops - test code was only testing failure of around 30 of the
2995 mallocs in the test path, but it turns out there are 500+
2996 mallocs. I believe this was due to misguided linking setup such
2997 that there was one copy of dbus_malloc etc. in the daemon and one
2998 in the shared lib, and only daemon mallocs were tested. In any
2999 case, the test case now tests all 500+ mallocs, and doesn't pass
3000 yet, though there are lots of fixes in this patch.
3002 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
3003 this so that it doesn't need to allocate memory, since it
3004 has no way of indicating failure due to OOM (and would be
3005 annoying if it did).
3007 * dbus/dbus-list.c (_dbus_list_pop_first_link): new function
3009 * bus/Makefile.am: rearrange to create two self-contained
3010 libraries, to avoid having libraries with overlapping symbols.
3011 that was resulting in weirdness, e.g. I'm pretty sure there
3012 were two copies of global static variables.
3014 * dbus/dbus-internals.c: move the malloc debug stuff to
3017 * dbus/dbus-list.c (free_link): free list mempool if it becomes
3020 * dbus/dbus-memory.c (_dbus_disable_mem_pools): new function
3022 * dbus/dbus-address.c (dbus_parse_address): free list nodes
3025 * bus/dispatch.c (bus_dispatch_add_connection): free
3026 message_handler_slot when no longer using it, so
3027 memory leak checkers are happy for the test suite.
3029 * dbus/dbus-server-debug-pipe.c (debug_finalize): free server name
3031 * bus/bus.c (new_connection_callback): disconnect in here if
3032 bus_connections_setup_connection fails.
3034 * bus/connection.c (bus_connections_unref): fix to free the
3036 (bus_connections_setup_connection): if this fails, don't
3037 disconnect the connection, just be sure there are no side
3040 * dbus/dbus-string.c (undo_alignment): unbreak this
3042 * dbus/dbus-auth.c (_dbus_auth_unref): free some stuff we were
3044 (_dbus_auth_new): fix the order in which we free strings
3047 * bus/connection.c (bus_connection_disconnected): fix to
3048 not send ServiceDeleted multiple times in case of memory
3051 * dbus/dbus-bus.c (dbus_bus_get_base_service): new function to
3052 get the base service name
3053 (dbus_bus_register_client): don't return base service name,
3054 instead store it on the DBusConnection and have an accessor
3056 (dbus_bus_register_client): rename dbus_bus_register()
3058 * bus/dispatch.c (check_hello_message): verify that other
3059 connections on the bus also got the correct results, not
3060 just the one sending hello
3062 2003-03-15 Havoc Pennington <hp@pobox.com>
3064 Make it pass the Hello handling test including all OOM codepaths.
3065 Now to do other messages...
3067 * bus/services.c (bus_service_remove_owner): fix crash when
3068 removing owner from an empty list of owners
3069 (bus_registry_ensure): don't leave service in the list of
3070 a connection's owned services if we fail to put the service
3073 * bus/connection.c (bus_connection_preallocate_oom_error): set
3074 error flag on the OOM error.
3076 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
3077 handle _dbus_transport_set_connection failure
3079 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd): modify
3080 to create watches up front and simply enable/disable them as
3082 (unix_connection_set): this can now fail on OOM
3084 * dbus/dbus-timeout.c, dbus/dbus-watch.c: add concept
3085 of enabling/disabling a watch or timeout.
3087 * bus/loop.c (bus_loop_iterate): don't touch disabled
3090 * glib/dbus-gmain.c: adapt to enable/disable watches and timeouts
3092 2003-03-15 Havoc Pennington <hp@pobox.com>
3094 * bus/dispatch.c (bus_dispatch_test): OK, now finally actually
3095 write useful test code, after all that futzing around ;-)
3097 Test does not yet pass because we can't handle OOM in
3098 _dbus_transport_messages_pending (basically,
3099 dbus_connection_preallocate_send() does not prealloc the write
3100 watch). To fix this, I think we need to add new stuff to
3101 set_watch_functions, namely a SetEnabled function so we can alloc
3102 the watch earlier, then enable it later.
3104 * dbus/Makefile.am (libdbus_convenience_la_SOURCES): move
3105 dbus-memory.c to the convenience lib
3107 * bus/test.c: rename some static functions to keep them clearly
3108 distinct from stuff in connection.c. Handle client disconnection.
3110 2003-03-14 Havoc Pennington <hp@pobox.com>
3112 * bus/dispatch.c (bus_dispatch_test): do test using debug-pipe
3113 transport, tests more of the real codepath. Set up clients
3114 with bus_setup_debug_client.
3116 * bus/test.c (bus_setup_debug_client): function to set up debug
3117 "clients" on the main loop
3119 * dbus/dbus-transport.c (_dbus_transport_open): add debug-pipe
3122 * dbus/dbus-server.c (dbus_server_listen): add debug-pipe
3125 * dbus/dbus-server-debug.c: support a debug server based on pipes
3127 * dbus/dbus-sysdeps.c (_dbus_full_duplex_pipe): new function
3128 (_dbus_close): new function
3130 * configure.in: check for socketpair
3132 2003-03-14 Havoc Pennington <hp@redhat.com>
3134 * dbus/dbus-memory.c: add a "detect buffer overwrites on free"
3137 * dbus/dbus-transport-debug.c: rework this a good bit to be
3138 less complicated. hopefully still works.
3140 * dbus/dbus-server-debug.c (handle_new_client): remove timeout
3143 * glib/dbus-gmain.c (timeout_handler): don't remove timeout
3146 * dbus/dbus-message.c (dbus_message_copy): rename from
3147 dbus_message_new_from_message, fix it up to copy
3148 all the message fields, add test case
3150 * bus/dispatch.c (bus_dispatch_test): add some more test code,
3151 not quite passing yet
3153 2003-03-14 Havoc Pennington <hp@pobox.com>
3155 * bus/loop.c (bus_loop_iterate): add this so we can "run loop
3156 until no work remains" in test code. (the large diff here
3157 is just code movement, no actual changes)
3159 * dbus/dbus-server-debug.c (DEFAULT_INTERVAL): change interval to
3160 1, no point waiting around for test code.
3161 (_dbus_server_debug_accept_transport): unref the timeout
3162 after adding it (right?)
3164 * dbus/dbus-transport-debug.c (DEFAULT_INTERVAL): ditto
3166 2003-03-13 Havoc Pennington <hp@redhat.com>
3168 * dbus/dbus-timeout.c (_dbus_timeout_list_set_functions): handle
3171 * dbus/dbus-watch.c (_dbus_watch_list_set_functions): handle out
3174 * dbus/dbus-connection.h: Make AddWatchFunction and
3175 AddTimeoutFunction return a bool so they can fail on out-of-memory
3177 * bus/bus.c (bus_context_new): set up timeout handlers
3179 * bus/connection.c (bus_connections_setup_connection): set up
3182 * glib/dbus-gmain.c: adapt to the fact that set_functions stuff
3185 * bus/bus.c (bus_context_new): adapt to changes
3187 * bus/connection.c: adapt to changes
3189 * test/watch.c: adapt to DBusWatch changes
3191 * bus/dispatch.c (bus_dispatch_test): started adding this but
3194 2003-03-14 Anders Carlsson <andersca@codefactory.se>
3196 * bus/dispatch.c (send_service_nonexistent_error): Fix typo.
3198 2003-03-13 Havoc Pennington <hp@pobox.com>
3200 * bus/test.c, bus/test.h, bus/Makefile.am, bus/test-main.c:
3201 set up a test framework as for the library
3203 2003-03-12 Havoc Pennington <hp@pobox.com>
3205 Throughout: purge global variables, introduce BusActivation,
3206 BusConnections, BusRegistry, etc. objects instead.
3208 * bus/bus.h, bus/bus.c: introduce BusContext as a global
3211 * test/Makefile.am (TEST_BINARIES): disable bus-test for now,
3212 going to redo this a bit differently I think
3214 2003-03-12 Havoc Pennington <hp@redhat.com>
3216 Mega-patch that gets the message bus daemon initially handling
3217 out-of-memory. Work still needed. Also lots of random
3218 moving stuff to DBusError instead of ResultCode.
3220 * dbus/dbus-list.c (_dbus_list_length_is_one): new function
3222 * dbus/dbus-connection.c
3223 (dbus_connection_send_with_reply_and_block): use DBusError
3225 * dbus/dbus-bus.c: adapt to API changes, make it use DBusError not
3228 * dbus/dbus-connection.c (dbus_connection_send): drop the result
3229 code here, as the only failure possible is OOM.
3231 * bus/connection.c (bus_connection_disconnect):
3232 rename bus_connection_disconnected as it's a notification only
3234 * bus/driver.c (bus_driver_handle_acquire_service): don't free
3235 "name" on get_args failure, should be done by get_args;
3236 don't disconnect client for bad args, just return an error.
3237 (bus_driver_handle_service_exists): ditto
3239 * bus/services.c (bus_services_list): NULL-terminate returned array
3241 * bus/driver.c (bus_driver_send_service_lost)
3242 (bus_driver_send_service_acquired): send messages from driver to a
3243 specific client to the client's unique name, not to the broadcast
3246 * dbus/dbus-message.c (decode_header_data): reject messages that
3247 contain no name field
3248 (_dbus_message_get_client_serial): rename to
3249 dbus_message_get_serial and make public
3250 (_dbus_message_set_serial): rename from set_client_serial
3251 (_dbus_message_set_reply_serial): make public
3252 (_dbus_message_get_reply_serial): make public
3254 * bus/connection.c (bus_connection_foreach): allow stopping
3255 iteration by returning FALSE from foreach function.
3257 * dbus/dbus-connection.c (dbus_connection_send_preallocated)
3258 (dbus_connection_free_preallocated_send)
3259 (dbus_connection_preallocate_send): new API for sending a message
3260 without possibility of malloc failure.
3261 (dbus_connection_send_message): rename to just
3262 dbus_connection_send (and same for whole function family)
3264 * dbus/dbus-errors.c (dbus_error_free): make this reinit the error
3266 * dbus/dbus-sysdeps.c (_dbus_exit): new function
3268 * bus/activation.c: handle/return errors
3270 * dbus/dbus-errors.h: add more DBUS_ERROR #define
3272 * dbus/dbus-sysdeps.c (_dbus_directory_open) (_dbus_file_get_contents)
3273 (_dbus_directory_get_next_file): use DBusError instead of DBusResultCode
3274 (_dbus_result_from_errno): move to this file
3276 2003-03-10 Anders Carlsson <andersca@codefactory.se>
3278 * dbus/dbus-marshal.c:
3279 (_dbus_marshal_set_string):
3280 Take a length argument so we can marshal the correct string
3283 (_dbus_marshal_dict), (_dbus_demarshal_dict),
3284 (_dbus_marshal_get_arg_end_pos), (_dbus_marshal_validate_arg),
3285 (_dbus_marshal_test):
3286 * dbus/dbus-marshal.h:
3287 Add support for marshalling and demarshalling dicts.
3289 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
3290 Add support for TYPE DICT.
3292 * dbus/dbus-message.c: (set_string_field):
3293 Adjust header padding.
3295 (dbus_message_append_args_valist), (dbus_message_append_dict),
3296 (dbus_message_get_args_valist), (dbus_message_iter_get_arg_type),
3297 (dbus_message_iter_get_dict), (_dbus_message_loader_return_buffer),
3298 (check_message_handling), (check_have_valid_message):
3299 * dbus/dbus-message.h:
3300 Add functions for setting and getting dicts.
3302 * dbus/dbus-protocol.h:
3308 * doc/dbus-specification.sgml:
3309 Add information about how dicts are marshalled.
3311 * test/data/invalid-messages/dict-with-nil-value.message:
3312 * test/data/invalid-messages/too-short-dict.message:
3313 * test/data/valid-messages/dict-simple.message:
3314 * test/data/valid-messages/dict.message:
3315 Add sample messages containing dicts.
3317 2003-03-08 Anders Carlsson <andersca@codefactory.se>
3319 * dbus/dbus-dict.h: Add DBUS_END_DECLS.
3321 2003-03-07 Anders Carlsson <andersca@codefactory.se>
3324 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
3325 (dbus_dict_get_keys), (insert_entry), (dbus_dict_set_boolean),
3326 (dbus_dict_set_int32), (dbus_dict_set_uint32),
3327 (dbus_dict_set_double), (dbus_dict_set_string),
3328 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
3329 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
3330 (dbus_dict_set_string_array), (_dbus_dict_test):
3332 Fix according to comments from Havoc.
3334 2003-03-06 Michael Meeks <michael@server.home>
3336 * configure.in: if we don't have kde-config, disable have_qt.
3338 2003-03-07 Anders Carlsson <andersca@codefactory.se>
3343 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
3344 (dbus_dict_ref), (dbus_dict_unref), (dbus_dict_contains),
3345 (dbus_dict_remove), (dbus_dict_get_value_type),
3346 (dbus_dict_get_keys), (dbus_dict_put_boolean),
3347 (dbus_dict_put_int32), (dbus_dict_put_uint32),
3348 (dbus_dict_put_double), (dbus_dict_put_string),
3349 (dbus_dict_put_boolean_array), (dbus_dict_put_int32_array),
3350 (dbus_dict_put_uint32_array), (dbus_dict_put_double_array),
3351 (dbus_dict_put_string_array), (dbus_dict_get_boolean),
3352 (dbus_dict_get_int32), (dbus_dict_get_uint32),
3353 (dbus_dict_get_double), (dbus_dict_get_string),
3354 (dbus_dict_get_boolean_array), (dbus_dict_get_int32_array),
3355 (dbus_dict_get_uint32_array), (dbus_dict_get_double_array),
3356 (dbus_dict_get_string_array), (_dbus_dict_test):
3358 Add DBusDict implementation
3360 * dbus/dbus-test.c: (dbus_internal_do_not_use_run_tests):
3364 2003-03-04 Havoc Pennington <hp@pobox.com>
3366 * test/data/auth/*: adapt to changes
3368 * dbus/dbus-auth-script.c (_dbus_auth_script_run): add
3369 USERID_BASE64 and change USERNAME_BASE64 to put in username not
3372 * dbus/dbus-keyring.c (_dbus_keyring_validate_context): prevent
3373 more stuff from being in a context name, to make the protocol
3374 simpler to deal with
3376 * dbus/dbus-errors.c (dbus_error_has_name): new function
3377 (dbus_error_is_set): new function
3379 * dbus/dbus-auth.c: replace DBUS_STUPID_TEST_MECH auth
3380 with DBUS_COOKIE_SHA1, implement DBUS_COOKIE_SHA1
3382 * dbus/dbus-connection.c (dbus_connection_flush): also read
3383 messages during a flush operation
3385 * dbus/Makefile.am: remove dbus-md5 since it isn't currently used.
3387 2003-03-05 Anders Carlsson <andersca@codefactory.se>
3389 * configure.in: Check for gethostbyname on Solaris.
3391 * dbus/dbus-transport.c: (_dbus_transport_open):
3392 Remove duplicate "tcp" entry.
3394 * doc/dbus-specification.sgml:
3395 Clarify some things.
3397 2003-03-05 Anders Carlsson <andersca@codefactory.se>
3399 * dbus/dbus-auth.c: (send_rejected), (process_test_subdir):
3400 * dbus/dbus-keyring.c: (_dbus_keyring_new_homedir),
3401 (_dbus_keyring_test):
3402 * dbus/dbus-md5.c: (_dbus_md5_compute):
3403 * dbus/dbus-sha.c: (_dbus_sha_compute):
3406 2003-03-05 Anders Carlsson <andersca@codefactory.se>
3408 * README: Add some things.
3410 2003-03-04 Anders Carlsson <andersca@codefactory.se>
3412 * dbus/dbus-message.c (dbus_message_append_args_valist): Add a break;
3413 after case DBUS_TYPE_BOOELAN.
3415 2003-03-02 Havoc Pennington <hp@pobox.com>
3417 * test/break-loader.c (randomly_set_extreme_ints): add test that
3418 sets really huge and small integers
3420 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add check
3421 that length of boolean array fits in the string, and that
3422 string has room for boolean value in single-bool case.
3424 * dbus/dbus-message-builder.c (_dbus_message_data_load): add
3425 optional value to "ALIGN" command which is what to fill the
3428 * test/data/valid-messages/no-padding.message: add regression
3429 test for the message padding problem
3431 2003-03-02 Havoc Pennington <hp@pobox.com>
3433 * dbus/dbus-message.c (decode_header_data): fix to always init
3434 message_padding, from Benjamin Dauvergne
3436 2003-03-02 Havoc Pennington <hp@pobox.com>
3442 2003-03-01 Joe Shaw <joe@assbarn.com>
3444 * configure.in: Check for "struct cmsgcred" and try to access its
3445 members for BSD-like unices.
3447 * dbus/dbus-sysdeps.c (read_credentials_byte): Fold this back into
3448 _dbus_read_credentials_unix_socket().
3449 (_dbus_read_credentials_unix_socket): Use recvmsg() instead of
3450 read() for reading the credential byte off the unix socket. Use
3451 struct cmsgcred on systems that support it.
3453 2003-02-27 Alexander Larsson <alexl@redhat.com>
3457 Make gthreads-2.0 dependency optional. Don't build thread test if
3460 2003-02-27 Havoc Pennington <hp@pobox.com>
3462 * dbus/dbus-connection.c
3463 (dbus_connection_send_message_with_reply_and_block): fix doh!
3464 doh! doh! bug that resulted in never removing a reply from the
3465 queue, no wonder we called get_reply_serial so much ;-)
3467 * dbus/dbus-message.c (struct DBusMessage): cache reply serial
3468 and client serial instead of demarshaling them every time
3470 2003-02-27 Havoc Pennington <hp@pobox.com>
3472 * dbus/dbus-marshal.c (_dbus_demarshal_int32): rewrite to be much
3473 more inlined, using dbus-string-private.h, speeds things up
3476 * dbus/dbus-string.c (_dbus_string_free): apply align offset
3477 when freeing the string
3478 (_dbus_string_steal_data): fix for align offset
3479 (undo_alignment): new function
3481 2003-02-26 Havoc Pennington <hp@redhat.com>
3483 All kinds of audit fixes from Owen, plus initial attempt to
3484 handle unaligned memory returned from malloc.
3486 * dbus/dbus-string.c (_dbus_string_init): clamp max length to
3487 leave room for align_offset and nul byte
3488 (fixup_alignment): function to track an align_offset and
3489 ensure real->str is aligned
3490 (DBUS_GENERIC_STRING_PREAMBLE): len must be less than allocated,
3491 to allow a nul byte plus align offset
3492 (_dbus_string_lock): fix overflow issue
3493 (_dbus_string_init_const_len): add assertions on sanity of len,
3494 assign allocated to be ALLOCATION_PADDING larger than len
3495 (set_length): fixup the overflow handling
3496 (_dbus_string_get_data_len): fix overflow in assertion
3497 (open_gap): detect overflow in size of gap to be opened
3498 (_dbus_string_lengthen): add overflow check
3499 (_dbus_string_align_length): fix overflow with _DBUS_ALIGN_VALUE
3500 (_dbus_string_append): add overflow check
3501 (_dbus_string_append_unichar): overflow
3502 (_dbus_string_delete): fix overflow in assertion
3503 (_dbus_string_copy_len): overflow in assertion
3504 (_dbus_string_replace_len): overflows in assertions
3505 (_dbus_string_find): change to implement in terms of
3506 _dbus_string_find_to
3507 (_dbus_string_find_to): assorted fixage
3508 (_dbus_string_equal_c_str): assert c_str != NULL,
3509 fix logic so the function works
3510 (_dbus_string_ends_with_c_str): fix overflow thingy
3511 (_dbus_string_base64_encode): overflow fix
3512 (_dbus_string_validate_ascii): overflow
3513 (_dbus_string_validate_nul): overflow
3515 2003-02-26 Havoc Pennington <hp@redhat.com>
3517 * dbus/dbus-marshal.c (_dbus_marshal_test): fix to work with DISABLE_ASSERTS
3519 2003-02-26 Alexander Larsson <alexl@redhat.com>
3522 Set DBUS_GLIB_THREADS_LIBS for apps using gthread-2.0
3524 * dbus/dbus-connection.c:
3525 * dbus/dbus-connection.h:
3526 Fix _dbus_connection_acquire_io_path and _dbus_connection_acquire_dispatch.
3527 Add dbus_connection_set_wakeup_main_function and use it when queueing
3528 incoming and outgoing messages.
3531 * dbus/dbus-dataslot.c:
3532 Threadsafe usage of DBusDataSlotAllocator
3534 * dbus/dbus-message.c: (dbus_message_get_args_iter):
3537 * dbus/dbus-server-unix.c:
3540 * glib/dbus-gmain.c:
3541 Implement the new wakeup functions for glib.
3544 * glib/test-thread-client.c:
3545 * glib/test-thread-server.c:
3546 * glib/test-thread.h:
3547 Initial cut at some thread test code. Not really done yet.
3549 2003-02-26 Havoc Pennington <hp@pobox.com>
3551 * dbus/dbus-connection.c
3552 (dbus_connection_send_message_with_reply_and_block): fix crash
3553 where we ref'd the outgoing message instead of the returned reply
3555 * dbus/dbus-transport-unix.c (do_authentication): check read watch
3556 at the end of this function, so if we didn't need to read for
3557 authentication, we reinstall it for receiving messages
3559 * dbus/dbus-message.c (dbus_message_new_reply): allow replies to
3560 a NULL sender for peer-to-peer case
3562 * dbus/dbus-transport-unix.c (check_read_watch): handle
3563 !authenticated case correctly
3565 * glib/dbus-gmain.c: add support for DBusServer
3567 * dbus/dbus-server.c: add data slot support
3569 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): check
3570 return values and handle errors
3572 * dbus/dbus-dataslot.c: factor out the data slot stuff from
3575 * Doxyfile.in (INPUT): add glib subdir
3577 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): rename
3578 setup_with_g_main instead of hookup_with_g_main; write docs
3580 2003-02-24 Anders Carlsson <andersca@codefactory.se>
3582 * dbus/dbus-marshal.c: (_dbus_marshal_validate_arg):
3583 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
3584 * dbus/dbus-message.c: (dbus_message_append_boolean),
3585 (dbus_message_append_boolean_array),
3586 (dbus_message_get_args_valist), (_dbus_message_test):
3587 * dbus/dbus-message.h:
3588 * doc/dbus-specification.sgml:
3589 Various fixes as pointed out by Havoc.
3591 * test/data/invalid-messages/bad-boolean-array.message:
3592 * test/data/invalid-messages/bad-boolean.message:
3593 Add invalid boolean value test cases.
3595 2003-02-24 Anders Carlsson <andersca@codefactory.se>
3597 * dbus/dbus-internals.c: (_dbus_type_to_string):
3598 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
3599 (_dbus_marshal_validate_arg):
3600 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
3601 * dbus/dbus-message.c: (dbus_message_append_args_valist),
3602 (dbus_message_append_boolean), (dbus_message_append_boolean_array),
3603 (dbus_message_get_args_valist), (dbus_message_iter_get_boolean),
3604 (dbus_message_iter_get_int32), (dbus_message_iter_get_uint32),
3605 (dbus_message_iter_get_double),
3606 (dbus_message_iter_get_boolean_array), (message_iter_test):
3607 * dbus/dbus-message.h:
3608 * dbus/dbus-protocol.h:
3609 * doc/dbus-specification.sgml:
3610 * test/data/valid-messages/lots-of-arguments.message:
3611 Add support for boolean and boolean array types.
3613 2003-02-23 Havoc Pennington <hp@pobox.com>
3615 * dbus/dbus-keyring.c: finish most of this implementation and
3618 * dbus/dbus-errors.c (dbus_set_error_const, dbus_set_error): make
3619 these barf if the error isn't cleared to NULL
3621 * dbus/dbus-sysdeps.c (_dbus_delete_file): set error on failure
3622 (_dbus_create_directory): new function
3624 * dbus/dbus-errors.c (dbus_set_error): fix warning
3626 * dbus/dbus-string.c (_dbus_string_hex_encode): new function
3627 (_dbus_string_hex_decode): new function
3628 (test_hex_roundtrip): test code
3630 * dbus/dbus-sha.c (_dbus_sha_compute): use dbus_string_hex_encode
3632 * dbus/dbus-md5.c (_dbus_md5_compute): use dbus_string_hex_encode
3634 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): make this use
3635 the save-to-temp/rename trick to atomically write the new file
3636 (_dbus_string_parse_uint): new function
3638 2003-02-22 Havoc Pennington <hp@pobox.com>
3640 * test/Makefile.am (dist-hook): fix dist for test/data/sha-1
3642 2003-02-22 Havoc Pennington <hp@pobox.com>
3644 * dbus/dbus-message.c (dbus_message_iter_get_string_array):
3645 (dbus_message_iter_get_byte_array): Fix up doxygen warnings
3647 * dbus/dbus-sha.c: add implementation of SHA-1 algorithm
3649 * dbus/test/data/sha-1: add US government test suite for SHA-1
3651 2003-02-21 Anders Carlsson <andersca@codefactory.se>
3653 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
3654 Make string arrays NULL-terminated.
3656 * dbus/dbus-memory.c: (dbus_free_string_array):
3657 * dbus/dbus-memory.h:
3658 New function for freeing NULL-terminated string arrays.
3660 * dbus/dbus-message-builder.c: (append_quoted_string),
3661 (_dbus_message_data_load):
3662 Add support for array types.
3664 * dbus/dbus-message.c: (check_message_handling):
3665 Add more types as test cases.
3667 * dbus/dbus-sysdeps.c: (_dbus_string_parse_int),
3668 (_dbus_string_parse_double):
3669 Add the start offset to the end offset.
3671 * test/data/valid-messages/lots-of-arguments.message:
3672 New test message with lots of arguments.
3674 2003-02-21 Anders Carlsson <andersca@codefactory.se>
3676 * dbus/dbus-message.c: (dbus_message_append_nil),
3677 (dbus_message_append_int32), (dbus_message_append_uint32),
3678 (dbus_message_append_double), (dbus_message_append_string),
3679 (dbus_message_append_int32_array),
3680 (dbus_message_append_uint32_array),
3681 (dbus_message_append_double_array),
3682 (dbus_message_append_byte_array),
3683 (dbus_message_append_string_array):
3684 Fix all out-of-memory handling in these functions.
3686 2003-02-21 Anders Carlsson <andersca@codefactory.se>
3688 * dbus/dbus-message.c: (dbus_message_append_nil):
3691 2003-02-21 Anders Carlsson <andersca@codefactory.se>
3693 * dbus/dbus-message.c: (dbus_message_append_args_valist),
3694 (dbus_message_append_nil), (dbus_message_append_int32_array),
3695 (dbus_message_append_uint32_array),
3696 (dbus_message_append_double_array),
3697 (dbus_message_append_byte_array),
3698 (dbus_message_append_string_array), (dbus_message_get_args_valist),
3699 (dbus_message_iter_get_int32_array),
3700 (dbus_message_iter_get_uint32_array),
3701 (dbus_message_iter_get_double_array),
3702 (dbus_message_iter_get_byte_array),
3703 (dbus_message_iter_get_string_array):
3705 * dbus/dbus-message.h:
3706 Add functions for appending and getting arrays.
3708 2003-02-21 Anders Carlsson <andersca@codefactory.se>
3710 * dbus/dbus-mempool.c (_dbus_mem_pool_new): Make the
3711 element size at least 8 bytes, fixes mempool tests on
3714 2003-02-20 Alexander Larsson <alexl@redhat.com>
3716 * dbus/dbus-transport-unix.c (unix_do_iteration):
3717 Unlock the connection mutex during a blocking select call.
3718 Add todo about how we need a way to wake up the select.
3720 * dbus/dbus-connection-internal.h:
3721 * dbus/dbus-connection.c:
3722 Add _dbus_connection_lock and _dbus_connection_unlock.
3724 2003-02-19 Havoc Pennington <hp@pobox.com>
3726 * Doxyfile.in (PREDEFINED): put DOXYGEN_SHOULD_SKIP_THIS in
3727 Doxyfile.in, not Doxyfile
3729 * dbus/dbus-keyring.c: do some hacking on this
3731 * dbus/dbus-sysdeps.c (_dbus_delete_file): new
3733 * dbus/dbus-errors.c (dbus_set_error_const): do not call
3735 (dbus_set_error): remove dbus_error_init, check for message ==
3736 NULL *before* we sprintf into it, and add @todo about including
3737 system headers in this file
3739 * dbus/dbus-sysdeps.c (_dbus_create_file_exclusively): new
3741 * dbus/dbus-errors.h (DBUS_ERROR_FAILED): add
3743 * dbus/dbus-sysdeps.c (get_user_info): break this function out to
3744 get various bits of user information based on either username
3746 (_dbus_homedir_from_username): new function
3748 2003-02-19 Anders Carlsson <andersca@codefactory.se>
3751 Add check for nonposix getpwnam_r
3753 * dbus/dbus-mempool.c: (_dbus_mem_pool_new):
3754 Align the pool element size to a sizeof (void *) boundary.
3756 * dbus/dbus-sysdeps.c: (_dbus_setenv), (_dbus_connect_unix_socket),
3757 (_dbus_listen_unix_socket), (_dbus_credentials_from_username):
3758 General Solaris fixes.
3760 * test/data/valid-messages/simplest-manual.message:
3761 Explicitly state that we want little-endian packing.
3763 2003-02-19 Mikael Hallendal <micke@codefactory.se>
3765 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
3767 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_tcp_socket):
3768 Added to create a transport connecting using a tcp/ip socket.
3770 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): Added to connect
3771 to a tcp socket at given host and port.
3772 (_dbus_listen_tcp_socket): added to listen on tcp socket for given
3775 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
3777 * dbus/dbus-server-unix.c (_dbus_server_new_for_tcp_socket):
3778 Added to create a server listening on a TCP/IP socket.
3780 2003-02-19 Havoc Pennington <hp@pobox.com>
3782 Throughout: mop up all the Doxygen warnings and undocumented
3785 * dbus/dbus-sysdeps.c (do_exec): do not use execvp, we don't want
3786 to search any paths.
3788 * dbus/dbus-threads.c: move global mutex initializers to
3789 dbus-internals.h, multiple prototypes was confusing doxygen
3790 besides being kind of ugly
3792 * Doxyfile (PREDEFINED): have Doxygen define
3793 DOXYGEN_SHOULD_SKIP_THIS so we can exclude things from
3794 docs with #ifndef DOXYGEN_SHOULD_SKIP_THIS
3795 (do not abuse the feature! it's for stuff like the autogenerated
3796 macros in dbus-md5.c, not just for things you don't feel like
3799 2003-02-18 Havoc Pennington <hp@pobox.com>
3801 * dbus/dbus-string.c (_dbus_string_zero): new function
3803 * dbus/dbus-md5.c: include MD5 implementation by L. Peter Deutsch,
3804 wrap it in some dbus-friendly API
3806 * dbus/dbus-types.h: add 16-bit types
3808 2003-02-18 Joe Shaw <joe@assbarn.com>
3810 * dbus/dbus-auth.c (handle_server_data_stupid_test_mech): Just get
3811 credentials from our currently running process.
3812 (get_word): Fix a buglet where we were copying the entire length
3813 instead of relative to our position.
3815 * dbus/dbus-hash.c (_dbus_hash_test): Don't try to allocate the
3816 keys on the stack... it's 640k of data.
3818 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): Always
3819 read the credentials byte off the socket, even if we don't have
3821 (_dbus_poll): Implement poll() using select() for systems which
3824 * glib/test-dbus-glib.c (main): Print out an error if no
3825 parameters are given.
3827 * test/data/auth/fallback.auth-script: Added. Tests that a client
3828 can fallback to a secondary auth mechanism if the first fails.
3830 2003-02-18 Havoc Pennington <hp@pobox.com>
3834 2003-02-17 Havoc Pennington <hp@pobox.com>
3836 * doc/dbus-specification.sgml: lots of cosmetic
3837 cleanups/rearrangement, add assorted FIXME, change DBUS_ADDRESS
3838 env variable to DBUS_BUS_ADDRESS, s/client/application/,
3839 s/server/bus/ (except in authentication section). Add a section
3840 "Message Bus Message Routing"
3842 2003-02-17 Anders Carlsson <andersca@codefactory.se.>
3848 2003-02-17 Anders Carlsson <andersca@codefactory.se>
3850 * doc/dbus-specification.sgml:
3851 Specification updates.
3853 2003-02-17 Anders Carlsson <andersca@codefactory.se>
3855 * bus/activation.c: (bus_activation_init), (child_setup),
3856 (bus_activation_activate_service):
3858 * bus/main.c: (main):
3859 Set DBUS_ADDRESS environment variable.
3861 * dbus/dbus-errors.c: (dbus_set_error):
3862 Don't use va_copy since that's a C99 feature.
3864 * dbus/dbus-sysdeps.c: (_dbus_setenv), (do_exec),
3865 (_dbus_spawn_async):
3866 * dbus/dbus-sysdeps.h:
3867 Add child_setup_func to _dbus_spawn_async.
3869 * doc/dbus-specification.sgml:
3870 Update specification.
3872 * test/spawn-test.c: (setup_func), (main):
3875 2003-02-17 Alexander Larsson <alexl@redhat.com>
3877 * dbus/dbus-connection.c (_dbus_connection_handler_destroyed_locked):
3880 2003-02-17 Anders Carlsson <andersca@codefactory.se>
3884 * doc/dbus-test-plan.sgml:
3885 Add test plan document.
3890 2003-02-17 Anders Carlsson <andersca@codefactory.se>
3892 * dbus/dbus-message.c: (decode_header_data),
3893 (_dbus_message_loader_return_buffer):
3894 Set the header padding amount when loading a message.
3896 2003-02-16 Anders Carlsson <andersca@codefactory.se>
3898 * bus/dispatch.c: (send_one_message):
3899 Only send broadcast messages to registered connections.
3901 * dbus/dbus-message.c: (dbus_message_name_is):
3902 * dbus/dbus-message.h:
3903 New convenience function.
3905 * dbus/dbus-transport-debug.c: (do_reading):
3906 Only dispatch one message per run.
3909 * test/bus-test.c: (new_connection_callback), (die),
3910 (test_hello_client1_handler), (test_hello_client2_handler),
3911 (test_hello_replies), (main):
3913 * test/bus-test-loop.[ch]:
3916 2003-02-16 Havoc Pennington <hp@pobox.com>
3918 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
3919 backward conditional
3921 2003-02-16 Alexander Larsson <alexl@redhat.com>
3923 * dbus/dbus-connection.c:
3924 Implement sent_message_with_reply. (with_reply_and block is still
3926 Made dispatch_message not lose message if OOM.
3928 * dbus/dbus-errors.h:
3929 Add NoReply error (for reply timeouts).
3931 2003-02-16 Alexander Larsson <alexl@redhat.com>
3933 * dbus/dbus-hash.c (_dbus_hash_table_unref):
3934 Actually free keys and values when destroying hashtable.
3936 2003-02-16 Anders Carlsson <andersca@codefactory.se>
3938 * dbus/dbus-auth.c: (client_try_next_mechanism):
3941 * dbus/dbus-threads.c: (dbus_condvar_wait_timeout):
3942 Return TRUE if there's no thread implementation around.
3944 * glib/dbus-gmain.c: (free_source),
3945 (dbus_connection_hookup_with_g_main):
3946 Make sure to remove the GSource when the connection is finalized.
3948 2003-02-16 Anders Carlsson <andersca@codefactory.se>
3950 * bus/dispatch.c: (bus_dispatch_message_handler):
3951 * dbus/dbus-errors.h:
3952 Return an error if someone tries to send a message to a service
3955 2003-02-16 Anders Carlsson <andersca@codefactory.se>
3957 * bus/activation.c: (load_directory), (bus_activation_init),
3958 (bus_activation_activate_service):
3961 (bus_driver_handle_activate_service), (bus_driver_handle_message):
3962 More work on the activation handling.
3964 * dbus/dbus-errors.h:
3965 Add some error messages
3967 * dbus/dbus-message.c: (dbus_message_new_error_reply):
3968 * dbus/dbus-message.h:
3969 New function that creates an error message.
3971 * dbus/dbus-protocol.h:
3972 Add ACTIVATE_SERVER message.
3974 * dbus/dbus-server-unix.c: (unix_handle_watch),
3975 (_dbus_server_new_for_domain_socket):
3976 Call _dbus_fd_set_close_on_exec.
3978 * dbus/dbus-sysdeps.c: (make_pipe), (do_exec),
3979 (_dbus_spawn_async), (_dbus_disable_sigpipe),
3980 (_dbus_fd_set_close_on_exec):
3981 * dbus/dbus-sysdeps.h:
3982 Add _dbus_fd_set_close_on exec function. Also add function that checks
3983 that all open fds are set to close-on-exec and warns otherwise.
3985 * dbus/dbus-transport-unix.c:
3986 (_dbus_transport_new_for_domain_socket):
3987 Call _dbus_fd_set_close_on_exec.
3989 2003-02-16 Havoc Pennington <hp@pobox.com>
3991 * dbus/dbus-connection.c (dbus_connection_set_change_sigpipe):
3992 allow people to avoid setting SIGPIPE to SIG_IGN
3993 (_dbus_connection_new_for_transport): disable SIGPIPE unless
3994 we've been asked not to
3996 2003-02-15 Anders Carlsson <andersca@codefactory.se>
3998 * dbus/dbus-list.c: (_dbus_list_append_link),
3999 (_dbus_list_prepend_link):
4000 * dbus/dbus-memory.c: (dbus_malloc), (dbus_malloc0),
4004 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4007 * bus/activation.c: (bus_activation_entry_free),
4008 (add_desktop_file_entry), (load_directory), (bus_activation_init):
4010 * bus/main.c: (main):
4011 Add simple activation support, doesn't work yet though.
4013 2003-02-15 Zack Rusin <zack@kde.org>
4015 * qt/dbus-qthread.cpp: small casting fix
4017 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4019 * dbus/dbus-errors.c: (dbus_set_error):
4020 * dbus/dbus-errors.h:
4021 Add a few errors and make dbus_set_error void.
4023 * dbus/dbus-sysdeps.c:
4024 (_dbus_errno_to_string), (close_and_invalidate), (make_pipe),
4025 (write_err_and_exit), (read_ints), (do_exec), (_dbus_spawn_async):
4026 * dbus/dbus-sysdeps.h:
4027 Add _dbus_spawn_async.
4029 * test/spawn-test.c: (main):
4030 Test for _dbus_spawn_async.
4032 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4034 * dbus/dbus-internals.h:
4035 Fix build without tests.
4037 * dbus/dbus-list.c: (alloc_link):
4038 Fix a segfault when a malloc fails.
4040 * dbus/dbus-memory.c: (initialize_malloc_debug), (dbus_malloc),
4041 (dbus_malloc0), (dbus_realloc):
4042 Add support for malloc debugging.
4044 2003-02-15 Alexander Larsson <alexl@redhat.com>
4046 * dbus/dbus-threads.c:
4047 * dbus/dbus-threads.h:
4048 Add condvars. Remove static mutext from API.
4049 Implement static mutexes by initializing them from threads_init.
4051 * glib/dbus-gthread.c:
4052 * qt/dbus-qthread.cpp:
4053 Update with the thread api changes.
4058 Turn StaticMutex into normal mutex + init function.
4059 Export new functions _dbus_list_alloc_link, _dbus_list_free_link,
4060 _dbus_list_append_link, _dbus_list_prepend_link
4063 * dbus/dbus-sysdeps.c:
4064 * dbus/dbus-sysdeps.h:
4065 New type dbus_atomic_t, and new functions _dbus_atomic_inc,
4066 _dbus_atomic_dec. Only slow fallback implementation at the moment.
4068 * dbus/dbus-protocol.h:
4069 Add DBUS_MESSAGE_LOCAL_DISCONNECT define
4071 * dbus/dbus-message.c:
4072 Make ref/unref atomic.
4075 * dbus/dbus-connection-internal.h:
4076 * dbus/dbus-connection.c:
4077 * dbus/dbus-connection.h:
4079 Change _peek to _borrow,_return & _steal_borrowed.
4080 Change disconnect callback to event.
4081 Make dbus_connection_dispatch_messages reentrant.
4083 * dbus/dbus-transport.c:
4084 Don't ref the connection on calls to the transport
4087 * dbus/dbus-message-handler.c:
4090 * glib/dbus-gmain.c:
4091 Don't use peek_message anymore
4094 * test/debug-thread.c:
4095 * test/debug-thread.h:
4096 Simple thread implementation that asserts() on deadlocks in
4097 single-threaded code.
4100 (main) Call debug_threads_init.
4103 Use disconnect message instead of disconnect callback.
4107 Don't call dbus_connection_set_disconnect_function. Instead export
4108 bus_connection_disconnect.
4111 Call bus_connection_disconnect when we get a disconnected message.
4113 2003-02-15 Havoc Pennington <hp@pobox.com>
4115 * dbus/dbus-message.c (dbus_message_new): fool around with the
4118 2003-02-14 Havoc Pennington <hp@pobox.com>
4120 * dbus/dbus-mempool.c: fail if the debug functions so indicate
4122 * dbus/dbus-memory.c: fail if the debug functions indicate we
4125 * dbus/dbus-internals.c (_dbus_set_fail_alloc_counter)
4126 (_dbus_decrement_fail_alloc_counter): debug functions to
4127 simulate memory allocation failures
4129 2003-02-14 Havoc Pennington <hp@pobox.com>
4131 * dbus/dbus-errors.h (struct DBusError): add a word of padding
4134 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4136 * bus/driver.c: (bus_driver_handle_hello):
4138 * bus/services.c: (bus_service_lookup):
4139 Reorder message sending so we get a more sane order.
4141 * test/bus-test.c: (message_handler):
4144 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4146 * bus/driver.c: (bus_driver_send_service_deleted),
4147 (bus_driver_send_service_created), (bus_driver_send_service_lost),
4148 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
4149 (bus_driver_send_welcome_message),
4150 (bus_driver_handle_list_services),
4151 (bus_driver_handle_acquire_service),
4152 (bus_driver_handle_service_exists):
4153 * dbus/dbus-bus.c: (dbus_bus_register_client),
4154 (dbus_bus_acquire_service), (dbus_bus_service_exists):
4155 * dbus/dbus-errors.c: (dbus_result_to_string):
4156 * dbus/dbus-errors.h:
4157 * dbus/dbus-message.c: (dbus_message_append_args),
4158 (dbus_message_append_args_valist), (dbus_message_get_args),
4159 (dbus_message_get_args_valist), (dbus_message_get_args_iter),
4160 (dbus_message_iter_get_arg_type), (dbus_message_iter_get_string),
4161 (dbus_message_iter_get_byte_array),
4162 (dbus_message_iter_get_string_array), (message_iter_test),
4163 (check_message_handling), (_dbus_message_test):
4164 * dbus/dbus-message.h:
4165 * test/bus-test.c: (main):
4166 Change fields to arguments in messages, so that they won't be
4167 confused with header fields.
4169 * glib/test-dbus-glib.c: (main):
4170 Remove append_fields from hello message.
4172 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4174 * dbus/dbus-errors.c:
4175 * dbus/dbus-message.c:
4176 * dbus/dbus-string.c:
4177 Documentation fixes.
4179 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4181 * glib/dbus-gmain.c: (timeout_handler), (add_timeout),
4183 Implement support for timeouts in dbus-glib.
4185 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4187 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
4188 * dbus/dbus-message.c: (process_test_subdir):
4189 * test/break-loader.c: (find_breaks_based_on):
4190 Plug some memory leaks.
4192 2003-02-13 Richard Hult <rhult@codefactory.se>
4194 * bus/main.c: Fix build.
4196 * dbus/dbus-errors.h:
4197 * dbus/dbus-errors.c: Fix copyright for Anders.
4199 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4204 * bus/connection.c: (bus_connection_foreach):
4207 * bus/desktop-file.c: (grow_lines_in_section), (grow_sections),
4208 (unescape_string), (new_section), (parse_section_start),
4209 (parse_key_value), (report_error), (bus_desktop_file_load),
4210 (bus_desktop_file_get_string):
4211 * bus/desktop-file.h:
4212 Use DBusError for error reporting.
4214 * bus/dispatch.c: (send_one_message),
4215 (bus_dispatch_message_handler):
4216 * bus/driver.c: (bus_driver_send_service_deleted),
4217 (bus_driver_send_service_created), (bus_driver_send_service_lost),
4218 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
4219 (bus_driver_send_welcome_message),
4220 (bus_driver_handle_list_services),
4221 (bus_driver_handle_acquire_service),
4222 (bus_driver_handle_service_exists):
4223 * bus/loop.c: (bus_loop_run):
4225 Use BUS_HANDLE_OOM instead of _DBUS_HANDLE_OOM.
4227 * bus/utils.c: (bus_wait_for_memory):
4229 New files with general utility functions.
4231 * dbus/dbus-internals.h:
4232 Remove _DBUS_HANDLE_OOM.
4234 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4236 * dbus/dbus-errors.c: (dbus_result_to_string), (dbus_error_init),
4237 (dbus_error_free), (dbus_set_error_const), (dbus_set_error):
4238 * dbus/dbus-errors.h:
4239 Add DBusError structure.
4241 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4243 * test/data/valid-messages/standard-acquire-service.message:
4244 * test/data/valid-messages/standard-hello.message:
4245 * test/data/valid-messages/standard-list-services.message:
4246 * test/data/valid-messages/standard-service-exists.message:
4247 Add some standard messages.
4249 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4251 * bus/driver.c: (bus_driver_send_welcome_message),
4252 (bus_driver_handle_list_services),
4253 (bus_driver_handle_acquire_service),
4254 (bus_driver_handle_service_exists), (bus_driver_handle_message):
4255 Update for API changes in libdbus.
4257 * dbus/dbus-message.c: (dbus_message_new_reply):
4258 * dbus/dbus-message.h:
4259 Remove the name argument. The spec states that replies shouldn't
4262 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4264 * bus/desktop-file.c: (parse_section_start), (parse_key_value),
4265 (report_error), (bus_desktop_file_load), (lookup_section),
4266 (lookup_line), (bus_desktop_file_get_raw),
4267 (bus_desktop_file_get_string):
4268 * bus/desktop-file.h:
4269 Some fixes, and new functions for getting a key value from a section.
4271 2003-02-13 Havoc Pennington <hp@pobox.com>
4273 * test/data/auth/fail-after-n-attempts.auth-script: new test
4275 * dbus/dbus-auth.c (send_rejected): shutdown_mech() when we
4278 2003-02-13 Havoc Pennington <hp@pobox.com>
4280 * dbus/dbus-auth.c (handle_server_data_external_mech): args to
4281 dbus_credentials_match were backward
4283 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support
4284 NO_CREDENTIALS and ROOT_CREDENTIALS
4286 * dbus/dbus-auth.c (_dbus_auth_do_work): move get_state() routine
4287 into here. Never process more commands after we've reached an
4288 end state; store further data as unused bytes.
4290 * test/data/auth/*: add more auth tests
4292 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support EXPECT
4293 command to match exact string and EXPECT_UNUSED to match unused
4296 * test/Makefile.am (dist-hook): fix to dist all the test stuff
4298 2003-02-12 Havoc Pennington <hp@pobox.com>
4300 * dbus/dbus-string.c (_dbus_string_pop_line): fix to also strip
4301 \r off of popped lines
4303 * dbus/dbus-auth.c (_dbus_auth_test): write code to run auth
4306 * dbus/dbus-auth-script.c (_dbus_auth_script_run): when doing a
4309 2003-02-12 Havoc Pennington <hp@pobox.com>
4311 * dbus/Makefile.am: remove break-loader from the build, since it
4314 * configure.in: add --enable-gcov to turn on coverage profiling
4315 flags and disable optimization
4317 2003-02-10 Havoc Pennington <hp@pobox.com>
4319 * dbus/dbus-auth-script.c, dbus/dbus-auth-script.h: sync
4320 initial cut at test framework for DBusAuth from laptop.
4321 Doesn't quite work yet but it compiles and I need to get
4322 it off the 266mhz laptop. ;-)
4324 * dbus/dbus-server-debug.c (_dbus_server_debug_accept_transport):
4325 fix a memleak in error case
4327 2003-02-12 Anders Carlsson <andersca@codefactory.se>
4330 * bus/desktop-file.c:
4331 * bus/desktop-file.h:
4332 Add a desktop file parser.
4334 2003-02-11 Zack Rusin <zack@kde.org>
4336 * qt/message.[h|cpp]: sample implementation
4337 of the KDE wrapper for DBusMessage
4339 2003-02-09 Zack Rusin <zack@kde.org>
4341 * test/bus-test.c: make_it_compile
4342 * doc/dbus-specification.sgml: minimal semantic fix
4344 2003-02-06 Anders Carlsson <andersca@codefactory.se>
4350 2003-02-06 Anders Carlsson <andersca@codefactory.se>
4353 * dbus/dbus-break-loader.c:
4355 * test/break-loader.c:
4356 Move dbus-break-loader to test/ and rename it to break-loader.
4358 2003-02-02 Havoc Pennington <hp@pobox.com>
4360 * dbus/dbus-keyring.c, dbus/dbus-keyring.h: template files
4361 for code to manage cookies in your home directory
4363 * dbus/dbus-sysdeps.c (_dbus_generate_random_bytes): new function
4365 * dbus/dbus-auth.c (get_state): impose a maximum number of tries
4366 to authenticate, then disconnect the client.
4368 2003-02-03 Alexander Larsson <alexl@redhat.com>
4370 * dbus/dbus-message.c (dbus_message_append_fields):
4373 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4375 * doc/dbus-specification.sgml:
4376 Update address format section.
4378 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4381 * test/bus-test.c: (get_time), (add_timeout), (remove_timeout),
4382 (message_handler), (new_connection_callback), (loop_quit),
4386 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4388 * bus/driver.c: (bus_driver_handle_service_exists):
4389 Simplify the code a bit.
4391 * dbus/dbus-bus.c: (dbus_bus_service_exists):
4394 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4397 Add libdbus-daemon.la and link to it.
4399 2003-02-01 James Willcox <jwillcox@gnome.org>
4401 * bus/driver.c: (bus_driver_handle_own_service):
4402 Actually include the service reply code in the message.
4404 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4406 * bus/driver.c: (bus_driver_handle_service_exists):
4407 Don't unref the incoming message.
4409 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4411 * dbus/dbus.h: Add dbus-address.h and dbus-bus.h
4413 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4415 * dbus/dbus-server.c: (dbus_server_listen):
4416 * dbus/dbus-transport.c: (_dbus_transport_open):
4417 ifdef out the calls to the debug transport and server.
4419 2003-02-02 Alexander Larsson <alexl@redhat.com>
4421 * dbus/dbus-watch.c (dbus_watch_get_flags):
4422 Add note in the docs that ERROR or HANGUP won't be returned
4423 and are assumed always on.
4425 * glib/dbus-gmain.c (add_watch):
4426 Always add IO_ERR | IO_HUP
4428 * dbus/dbus-message.h:
4429 Add semicolon after dbus_message_iter_get_string_array().
4430 Makes qt code build again
4432 2003-02-01 Anders Carlsson <andersca@codefactory.se>
4434 * bus/driver.c: (create_unique_client_name),
4435 (bus_driver_handle_hello):
4436 Don't take a name, just use a numeric id to identify
4440 * dbus/dbus-bus.c: (dbus_bus_register_client),
4441 (dbus_bus_acquire_service), (dbus_bus_service_exists):
4443 Add new convenience functions for communicating with the bus.
4445 * dbus/dbus-message.h:
4447 * dbus/dbus-protocol.h:
4450 2003-02-01 Alexander Larsson <alexl@redhat.com>
4452 * dbus/dbus-message.c (dbus_message_append_fields):
4453 Add some more doc comments.
4455 2003-02-01 Havoc Pennington <hp@pobox.com>
4457 * dbus/dbus-break-loader.c (randomly_modify_length): change
4458 a 4-byte value in the message as if it were a length
4460 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): don't set
4461 execute bit on saved files
4463 2003-02-01 Havoc Pennington <hp@pobox.com>
4465 * dbus/dbus-break-loader.c (main): new program to find messages
4466 that break the loader.
4468 * dbus/dbus-sysdeps.c (_dbus_string_append_uint): new function
4469 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): new function
4471 * dbus/dbus-string.c (_dbus_string_set_byte): new
4473 2003-01-31 Havoc Pennington <hp@pobox.com>
4475 * dbus/dbus-message.c: refactor the test code to be more general,
4476 in preparation for writing a "randomly permute test cases to
4477 try to break the loader" program.
4479 2003-01-31 Havoc Pennington <hp@pobox.com>
4481 * doc/dbus-specification.sgml: work on the specification
4483 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): check
4484 the protocol version of the message.
4486 * dbus/dbus-protocol.h: drop special _REPLY names, the spec
4487 no longer specifies that.
4488 (DBUS_SERVICE_REPLY_SERVICE_EXISTS): fix flags (1/2/4/8 not
4491 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos): add missing
4492 "break" for DBUS_TYPE_NIL, remove @todo
4494 2003-01-31 Havoc Pennington <hp@pobox.com>
4496 * dbus/dbus-message.c (dbus_message_set_is_error_reply): rename
4497 just set_is_error/get_is_error as this is a commonly-used
4498 function, and write docs.
4500 2003-01-31 Anders Carlsson <andersca@codefactory.se>
4502 * dbus/dbus-address.c: (dbus_address_entry_free):
4503 Free key and value lists.
4505 * dbus/dbus-internals.c: (_dbus_type_to_string):
4506 Add the types we didn't have.
4508 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
4509 (_dbus_marshal_validate_arg):
4512 * dbus/dbus-message.c: (dbus_message_set_sender):
4513 Remove todo about being able to set sender to NULL.
4515 (dbus_message_set_is_error_reply),
4516 (dbus_message_get_is_error_reply):
4517 * dbus/dbus-message.h:
4520 * dbus/dbus-protocol.h:
4521 Add error reply flag.
4523 * test/data/valid-messages/opposite-endian.message:
4524 Add NIL type to test.
4526 2003-01-31 Havoc Pennington <hp@pobox.com>
4528 * doc/dbus-specification.sgml: fully specify the header. Add
4529 flags and major protocol version, and change header/body len to
4532 * dbus/dbus-message-builder.c (append_saved_length): append length
4535 * dbus/dbus-message.c (dbus_message_create_header): change header
4536 length and body length to unsigned. Add the new fields from the
4538 (_dbus_message_loader_return_buffer): unsigned header/body len
4540 2003-01-30 Havoc Pennington <hp@pobox.com>
4542 * dbus/dbus-auth.c: rework to use only REJECTED, no
4545 * doc/dbus-sasl-profile.txt: drop MECHANISMS and just
4546 use REJECTED, suggested by Mark McLoughlin
4548 2003-01-30 Havoc Pennington <hp@pobox.com>
4550 * dbus/dbus-server.c (dbus_server_listen): @todo about how we need
4551 a better way to report errors here. e.g. "unix address lacks
4552 path" or something. also "no such file" when the path doesn't
4555 * dbus/dbus-address.c (dbus_address_entries_free): add @todo about
4557 (dbus_parse_address): add @todo about documenting address format,
4558 and allowing , and ; to be escaped
4560 2003-01-30 Anders Carlsson <andersca@codefactory.se>
4563 Add dbus-address.[ch]
4565 * dbus/dbus-address.c: (dbus_address_entry_free),
4566 (dbus_address_entries_free), (create_entry),
4567 (dbus_address_entry_get_method), (dbus_address_entry_get_value),
4568 (dbus_parse_address), (_dbus_address_test):
4569 * dbus/dbus-address.h:
4570 New files for dealing with address parsing.
4572 * dbus/dbus-connection.c:
4573 Document timeout functions.
4575 * dbus/dbus-message.c:
4576 Document dbus_message_new_from_message.
4578 * dbus/dbus-server-debug.c:
4581 * dbus/dbus-server.c: (dbus_server_listen):
4582 Parse address and use correct server implementation.
4584 * dbus/dbus-string.c: (_dbus_string_find_to), (_dbus_string_test):
4585 * dbus/dbus-string.h:
4586 New function with test.
4588 * dbus/dbus-test.c: (dbus_internal_symbol_do_not_use_run_tests):
4592 * dbus/dbus-transport-debug.c:
4595 * dbus/dbus-transport.c: (_dbus_transport_open):
4596 Parse address and use correct transport implementation.
4598 2003-01-30 Havoc Pennington <hp@pobox.com>
4600 * dbus/dbus-message.c: use message->byte_order instead of
4601 DBUS_COMPILER_BYTE_ORDER throughout.
4602 (dbus_message_create_header): pad header to align the
4603 start of the body of the message to 8-byte boundary
4605 * dbus/dbus-marshal.h: make all the demarshalers take const
4606 DBusString arguments.
4608 * dbus/dbus-message.c (_dbus_message_loader_return_buffer):
4609 validate message args here, so we don't have to do slow validation
4610 later, and so we catch bad messages as they are incoming. Also add
4611 better checks on header_len and body_len. Also fill in
4614 * dbus/dbus-string.c (_dbus_string_validate_utf8): new (not
4615 implemented properly)
4616 (_dbus_string_validate_nul): new function to check all-nul
4618 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): rename
4619 get_arg_end_pos and remove all validation
4620 (_dbus_marshal_validate_arg): actually do validation here.
4622 2003-01-29 Havoc Pennington <hp@pobox.com>
4624 * dbus/dbus-message.c (check_message_handling): fix assertion
4625 failure on set_client_serial
4627 2003-01-28 Havoc Pennington <hp@pobox.com>
4629 * dbus/dbus-server-debug.c: Add doc section comments
4631 * dbus/dbus-transport-debug.c: add doc section comments
4633 2003-01-28 Havoc Pennington <hp@redhat.com>
4635 * dbus/dbus-string.c (_dbus_string_base64_decode): append bytes in
4636 the reverse order from how I had it
4637 (_dbus_string_base64_encode): reverse encoding order. I was
4638 basically byteswapping everything during encoding.
4640 2003-01-28 Anders Carlsson <andersca@codefactory.se>
4642 * dbus/dbus-connection-internal.h:
4643 * dbus/dbus-connection.c: (_dbus_connection_add_timeout),
4644 (_dbus_connection_remove_timeout):
4645 Add functions for adding and removing timeouts.
4647 * dbus/dbus-message.c: (dbus_message_new_from_message):
4648 Add new function that takes a message and creates an exact
4649 copy of it, but with the refcount set to 1.
4650 (check_message_handling):
4653 * dbus/dbus-server-protected.h:
4654 * dbus/dbus-server.c: (_dbus_server_init_base),
4655 (_dbus_server_finalize_base), (_dbus_server_add_timeout),
4656 (dbus_server_set_timeout_functions):
4657 (_dbus_server_remove_timeout):
4658 New functions so that a server can add and remove timeouts.
4660 (dbus_server_listen):
4661 Add commented out call to dbus_server_debug_new.
4663 * dbus/dbus-timeout.c: (_dbus_timeout_new):
4664 Actually set the handler, doh.
4666 * dbus/dbus-transport.c: (_dbus_transport_open):
4667 Add commented out call to dbus_transport_debug_client_new.
4670 Add dbus-transport-debug.[ch] and dbus-server-debug.[ch]
4672 2003-01-28 Havoc Pennington <hp@pobox.com>
4674 * dbus/dbus-message.c (check_message_handling): function to check
4675 on the loaded message, iterates over it etc.
4677 2003-01-28 Havoc Pennington <hp@pobox.com>
4679 * test/Makefile.am (dist-hook): fix make distdir
4681 * dbus/Makefile.am (TESTS_ENVIRONMENT): fix make check
4683 2003-01-27 Havoc Pennington <hp@pobox.com>
4685 * dbus/dbus-mempool.c (time_for_size): replace printf with
4688 * dbus/dbus-message-builder.c (_dbus_message_data_load): allow
4689 empty lines; fix the SAVE_LENGTH stuff to be
4690 START_LENGTH/END_LENGTH so it actually works; couple other
4693 * test/Makefile.am (dist-hook): add dist-hook for .message files
4695 * dbus/dbus-string.c (DBUS_STRING_COPY_PREAMBLE): source of a copy
4696 can be constant or locked.
4697 (_dbus_string_free): allow freeing a const string as
4700 * dbus/dbus-sysdeps.c (_dbus_concat_dir_and_file): utility
4702 * dbus/dbus-test-main.c (main): take an argument which is the
4703 directory containing test data
4705 * dbus/dbus-message.c (_dbus_message_test): pass a test_data_dir
4706 argument to this and load all the messages in test/data/
4707 checking that they can be loaded or not loaded as appropriate.
4709 2003-01-27 Anders Carlsson <andersca@codefactory.se>
4711 * bus/dispatch.c: (bus_dispatch_message_handler):
4712 Dispatch messages sent to services.
4714 * bus/driver.c: (bus_driver_send_service_deleted),
4715 (bus_driver_send_service_created), (bus_driver_send_service_lost),
4716 (bus_driver_send_service_acquired):
4717 Add helper functions for sending service related messages.
4719 (bus_driver_send_welcome_message):
4720 Send HELLO_REPLY instead of WELCOME.
4722 (bus_driver_handle_list_services):
4723 Send LIST_SERVICES_REPLY instead of SERVICES.
4725 (bus_driver_handle_own_service),
4726 (bus_driver_handle_service_exists):
4727 New message handlers.
4729 (bus_driver_handle_message):
4730 Invoke new message handlers.
4732 (bus_driver_remove_connection):
4733 Don't remove any services here since that's done automatically
4734 by bus_service_remove_owner now.
4737 New function signatures.
4739 * bus/services.c: (bus_service_add_owner):
4740 Send ServiceAcquired message if we're the only primary owner.
4742 (bus_service_remove_owner):
4743 Send ServiceAcquired/ServiceLost messages.
4745 (bus_service_set_prohibit_replacement),
4746 (bus_service_get_prohibit_replacement):
4747 Functions for setting prohibit replacement.
4749 (bus_service_has_owner):
4750 New function that checks if a connection is in the owner queue of
4754 Add new function signatures.
4756 * dbus/dbus-list.c: (_dbus_list_test):
4757 Add tests for _dbus_list_remove_last and traversing the list backwards.
4760 Fix a typo in _dbus_list_get_prev_link, if we're at the first element we can't
4761 go any further, so return NULL then.
4763 * dbus/dbus-protocol.h:
4764 Add new messages, service flags and service replies.
4766 2003-01-26 Havoc Pennington <hp@pobox.com>
4768 * dbus/dbus-message-builder.c: implement, completely untested.
4770 * test/data/*: add data to be used in testing.
4771 ".message" files are our simple loadable text format.
4772 ".message-raw" will be binary dumps of messages.
4774 * dbus/dbus-string.c (_dbus_string_starts_with_c_str): new
4776 2003-01-26 Havoc Pennington <hp@pobox.com>
4778 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): new function
4780 * dbus/dbus-errors.c (dbus_result_to_string): add
4783 * dbus/dbus-message-builder.c: new file, will contain code to load
4784 up messages from files. Not implemented yet.
4786 2003-01-26 Havoc Pennington <hp@pobox.com>
4788 * dbus/dbus-message.c (dbus_message_set_sender): support deleting
4789 the sender by setting to NULL
4791 2003-01-26 Havoc Pennington <hp@pobox.com>
4793 The unit tests pass, but otherwise untested. If it breaks, the
4794 tests should have been better. ;-)
4796 * bus/driver.c (bus_driver_handle_hello): return if we disconnect
4799 * dbus/dbus-message.c: redo everything so we maintain
4800 message->header as the only copy of the various fields.
4801 This avoids the possibility of out-of-memory in some cases,
4802 for example dbus_message_lock() can't run out of memory anymore,
4803 and avoids extra copying. Figured I may as well go ahead and do
4804 this since it was busted for dbus_message_lock to not return
4805 failure on OOM, and dbus_message_write_header was totally
4806 unchecked for OOM. Also fixed some random other bugs.
4808 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): verify
4809 that strings are nul-terminated. Also, end_pos can be equal
4810 to string length just not greater than, I think.
4811 (_dbus_marshal_set_int32): new function
4812 (_dbus_marshal_set_uint32): new function
4813 (_dbus_marshal_set_string): new function
4815 * dbus/dbus-connection.c (_dbus_connection_new_for_transport): fix
4816 a warning, init timeout_list to NULL
4817 (dbus_connection_send_message): don't use uninitialized variable
4820 * dbus/dbus-string.c (_dbus_string_replace_len): new function
4822 2003-01-26 Anders Carlsson <andersca@codefactory.se>
4824 * bus/driver.c: (bus_driver_handle_hello),
4825 (bus_driver_send_welcome_message):
4828 2003-01-26 Anders Carlsson <andersca@codefactory.se>
4830 * dbus/dbus-auth.c: (process_auth), (_dbus_auth_unref):
4831 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
4832 (dbus_connection_unref):
4833 * dbus/dbus-marshal.c: (_dbus_marshal_test):
4834 * dbus/dbus-message.c: (dbus_message_unref),
4837 (dbus_message_get_fields):
4838 Remove debugging printout.
4840 (_dbus_message_loader_return_buffer):
4841 Don't store the header string.
4843 (_dbus_message_test):
4846 2003-01-26 Richard Hult <rhult@codefactory.se>
4848 * glib/dbus-gmain.c (dbus_connection_dispatch): Traverse a copy of
4849 the file descriptor list, since it can change under us.
4851 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4853 * glib/dbus-gmain.c: (dbus_connection_prepare),
4854 (dbus_connection_check), (dbus_connection_dispatch), (add_watch),
4855 (remove_watch), (dbus_connection_hookup_with_g_main):
4856 Rewrite the glib handling to use its own GSource instead of a
4857 GIOChannel so we can catch messages put in the queue while waiting
4860 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4863 * bus/connection.c: (connection_disconnect_handler),
4864 (connection_watch_callback), (bus_connection_setup):
4865 * bus/dispatch.c: (send_one_message),
4866 (bus_dispatch_broadcast_message), (bus_dispatch_message_handler),
4867 (bus_dispatch_add_connection), (bus_dispatch_remove_connection):
4869 * bus/driver.c: (bus_driver_send_service_deleted),
4870 (bus_driver_send_service_created), (bus_driver_handle_hello),
4871 (bus_driver_send_welcome_message),
4872 (bus_driver_handle_list_services), (bus_driver_remove_connection),
4873 (bus_driver_handle_message):
4875 Refactor code, put the message dispatching in its own file. Use
4876 _DBUS_HANDLE_OOM. Also send ServiceDeleted messages when a client
4879 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4881 * dbus/dbus-internals.h:
4882 Add _DBUS_HANDLE_OOM macro, it doesn't do anything currently.
4884 * dbus/dbus-message.c: (dbus_message_get_sender):
4885 * dbus/dbus-message.h:
4886 Implement dbus_message_get_sender.
4888 * dbus/dbus-protocol.h:
4889 Add message and service defines.
4891 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4893 * dbus/dbus-connection.c: (dbus_connection_send_message):
4894 * dbus/dbus-message-internal.h:
4895 * dbus/dbus-message.c: (_dbus_message_get_client_serial),
4896 (dbus_message_write_header):
4897 Remove _dbus_messag_unlock and don't set the client serial on a
4898 message if one already exists.
4900 2003-01-24 Havoc Pennington <hp@pobox.com>
4902 * dbus/dbus-list.c (alloc_link): put a thread lock on the global
4905 * bus/driver.c (bus_driver_handle_list_services): fix a leak
4908 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4910 * dbus/dbus-list.c: (alloc_link), (free_link):
4911 Use a memory pool for the links.
4913 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4915 * bus/connection.c: (bus_connection_foreach):
4917 Add new bus_connection_foreach function.
4919 * bus/driver.c: (send_one_message), (bus_driver_broadcast_message):
4920 Add function that broadcasts a message to all clients.
4922 (bus_driver_send_service_created), (bus_driver_handle_hello),
4923 (bus_driver_send_welcome_message),
4924 (bus_driver_handle_list_services), (bus_driver_message_handler):
4925 Implement functions that take care of listing services, and notifying
4926 clients when new services are created.
4928 * bus/services.c: (bus_services_list):
4930 Add new function that returns an array of strings with the currently
4931 registered services.
4934 * glib/dbus-gmain.c:
4935 Update copyright year.
4937 2003-01-25 Anders Carlsson <andersca@codefactory.se>
4939 * dbus/dbus-connection.c: (dbus_connection_send_message):
4940 Unlock the message in case it was sent earlier.
4942 (dbus_connection_send_message_with_reply_and_block):
4943 Remove the reply message from the list.
4945 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
4946 Set array_len and new_pos correctly.
4948 (_dbus_marshal_test):
4949 Remove debug output.
4951 * dbus/dbus-message-internal.h:
4952 * dbus/dbus-message.c: (_dbus_message_get_reply_serial):
4953 New function that returns the reply serial.
4955 (_dbus_message_unlock):
4956 New function that unlocks a message and resets its header.
4958 (dbus_message_append_string_array),
4959 (dbus_message_get_fields_valist),
4960 (dbus_message_iter_get_field_type),
4961 (dbus_message_iter_get_string_array),
4962 (dbus_message_get_fields),
4963 (dbus_message_append_fields_valist):
4964 Handle string arrays.
4966 (dbus_message_set_sender):
4967 Make this function public since the bus daemon needs it.
4969 (decode_header_data):
4970 Set the reply serial to -1 initially.
4972 * dbus/dbus-message.h:
4973 Add dbus_message_set_sender.
4975 2003-01-24 Havoc Pennington <hp@pobox.com>
4977 * doc/dbus-specification.sgml: add some stuff
4979 2003-01-22 Havoc Pennington <hp@pobox.com>
4981 * doc/dbus-specification.sgml: Start to document the protocol.
4983 2003-01-22 Havoc Pennington <hp@pobox.com>
4985 * dbus/dbus-connection.c
4986 (dbus_connection_send_message_with_reply_and_block): add some @todo
4988 * bus/driver.c (bus_driver_add_connection): add a FIXME about memleak
4990 2003-01-21 Havoc Pennington <hp@pobox.com>
4992 (patch untested because can't compile)
4994 * bus/driver.c (create_unique_client_name): make this function
4995 never recycle client names. Also, caller should initialize
4998 * dbus/dbus-sysdeps.c (_dbus_get_current_time): new function
5000 2003-01-21 Anders Carlsson <andersca@codefactory.se>
5002 * dbus/dbus-marshal.c: (_dbus_marshal_double),
5003 (_dbus_marshal_int32), (_dbus_marshal_uint32),
5004 (_dbus_marshal_int32_array), (_dbus_marshal_uint32_array),
5005 (_dbus_marshal_double_array), (_dbus_marshal_string_array),
5006 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
5007 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
5008 (_dbus_marshal_get_field_end_pos), (_dbus_marshal_test):
5009 * dbus/dbus-marshal.h:
5010 * dbus/dbus-protocol.h:
5011 Add support for marshalling and demarshalling integer, double
5014 2003-01-21 Anders Carlsson <andersca@codefactory.se>
5019 * bus/connection.c: (connection_disconnect_handler):
5020 Remove the connection from the bus driver's list.
5022 (connection_watch_callback): Dispatch messages.
5024 (free_connection_data): Free connection name.
5026 (bus_connection_setup): Add connection to the bus driver's list.
5027 (bus_connection_remove_owned_service):
5028 (bus_connection_set_name), (bus_connection_get_name):
5029 Add functions for setting and getting the connection's name.
5032 Add function headers.
5034 * bus/driver.c: (create_unique_client_name),
5035 (bus_driver_handle_hello_message),
5036 (bus_driver_send_welcome_message), (bus_driver_message_handler),
5037 (bus_driver_add_connection), (bus_driver_remove_connection):
5040 * bus/services.c: (bus_service_free):
5042 New file that handles communication and registreation with the bus
5045 2003-01-21 Anders Carlsson <andersca@codefactory.se>
5047 * dbus/dbus-connection.c: (dbus_connection_send_message):
5048 Add a new client_serial parameter.
5050 (dbus_connection_send_message_with_reply):
5051 Remove a @todo since we've implemented the blocking function.
5053 (dbus_connection_send_message_with_reply_and_block):
5054 New function that sends a message and waits for a reply and
5055 then returns the reply.
5057 * dbus/dbus-connection.h:
5060 * dbus/dbus-errors.c: (dbus_result_to_string):
5061 * dbus/dbus-errors.h:
5062 Add new DBUS_RESULT.
5064 * dbus/dbus-message-internal.h:
5065 * dbus/dbus-message.c: (_dbus_message_get_reply_serial),
5066 (_dbus_message_set_sender), (dbus_message_write_header),
5067 (dbus_message_new_reply), (decode_header_data),
5068 (_dbus_message_loader_return_buffer), (_dbus_message_test):
5069 * dbus/dbus-message.h:
5070 Add new functions that set the reply serial and sender.
5071 Also marshal and demarshal them correctly and add test.
5073 * dbus/dbus-protocol.h:
5074 Add new DBUS_MESSAGE_TYPE_SENDER.
5077 * glib/dbus-gmain.c: (watch_callback), (free_callback_data),
5078 (add_watch), (remove_watch), (add_timeout), (remove_timeout),
5079 (dbus_connection_hookup_with_g_main):
5080 * glib/test-dbus-glib.c: (main):
5081 Rewrite to use GIOChannel and remove the GSource crack.
5083 * test/echo-client.c: (main):
5084 * test/watch.c: (check_messages):
5085 Update for changed APIs
5087 2003-01-19 Anders Carlsson <andersca@codefactory.se>
5089 * dbus/Makefile.am: Add dbus-timeout.[cħ]
5091 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport):
5092 Create a DBusTimeoutList.
5093 (dbus_connection_set_timeout_functions): Add new function to
5094 set timeout callbacks
5096 * dbus/dbus-connection.h: Add public DBusTimeout API.
5098 * dbus/dbus-message.c: (dbus_message_get_service):
5099 * dbus/dbus-message.h: New function.
5101 * dbus/dbus-server.c: Fix small doc typo.
5103 * dbus/dbus-timeout.[ch]: New files for mainloop timeouts.
5105 2003-01-19 Anders Carlsson <andersca@codefactory.se>
5107 * dbus/dbus-string.c (_dbus_string_move_len): Don't delete all
5108 of the string, just as long as specified.
5110 2003-01-19 Havoc Pennington <hp@pobox.com>
5112 * dbus/dbus-connection.c (dbus_connection_get_is_authenticated):
5115 * dbus/dbus-server.c (dbus_server_set_max_connections)
5116 (dbus_server_get_max_connections, dbus_server_get_n_connections):
5117 keep track of current number of connections, and add API for
5118 setting a max (but haven't implemented enforcing the max yet)
5120 2003-01-18 Havoc Pennington <hp@pobox.com>
5122 * dbus/dbus-transport-unix.c (unix_do_iteration): only do the
5123 reading/writing if read_watch != NULL or write_watch != NULL.
5125 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): fix
5126 the message loader code to actually load message->header and
5127 message->body into the newly-created message.
5129 * dbus/dbus-transport-unix.c (check_write_watch): fix a mem leak
5132 * dbus/dbus-connection.c (dbus_connection_set_max_message_size)
5133 (dbus_connection_get_max_message_size)
5134 (dbus_connection_set_max_live_messages_size)
5135 (dbus_connection_get_max_live_messages_size): implement some
5136 resource limitation functions
5138 * dbus/dbus-resources.c: new file implementing some of the
5139 resource limits stuff
5141 * dbus/dbus-message.c (dbus_message_iter_get_byte_array): add
5142 missing docs, add @todo to handle OOM etc.
5144 * dbus/dbus-marshal.c (_dbus_demarshal_byte_array): add missing
5147 2003-01-18 Havoc Pennington <hp@pobox.com>
5149 * dbus/dbus-connection.c (dbus_connection_unref): disconnect the
5150 connection if it hasn't been already.
5152 * dbus/dbus-connection.h: kill off the idea of an ErrorFunction,
5153 replace with DisconnectFunction.
5155 2003-01-18 Havoc Pennington <hp@pobox.com>
5157 Building --disable-verbose-mode --disable-asserts --disable-tests
5158 cuts the library from 112K to 45K or so
5160 * configure.in: check for varargs macro support,
5161 add --enable-verbose-mode, --enable-asserts.
5163 * dbus/dbus-internals.h (_dbus_assert): support
5165 (_dbus_verbose): support DBUS_ENABLE_VERBOSE_MODE
5167 2003-01-18 Havoc Pennington <hp@pobox.com>
5169 * dbus/dbus-test.c: include config.h so that tests actually run
5171 * dbus/dbus-string.c: add assertions that stuff is 8-byte aligned,
5172 so the failure mode when that assumption fails will be plenty
5175 2003-01-18 Havoc Pennington <hp@pobox.com>
5177 * configure.in: default --enable-tests to $USE_MAINTAINER_MODE
5179 * dbus/Makefile.am: fix it up so dubs-test-main.c is included in
5182 * test/Makefile.am: don't use special variable "TESTS" for echo-*
5183 since we don't want to use those in make check
5185 2003-01-15 Havoc Pennington <hp@redhat.com>
5191 2003-01-15 Havoc Pennington <hp@redhat.com>
5193 * test/Makefile.am: fix so that test source code ends up in the
5194 distribution on make distcheck
5196 2003-01-15 Havoc Pennington <hp@redhat.com>
5202 2003-01-15 Havoc Pennington <hp@redhat.com>
5204 * dbus/dbus-test.c (dbus_internal_symbol_do_not_use_run_tests):
5205 fix build when --disable-tests
5207 * Makefile.am (EXTRA_DIST): put HACKING in here
5209 * HACKING: document procedure for making a tarball release.
5211 2003-01-14 Anders Carlsson <andersca@codefactory.se>
5213 * bus/connection.c: (connection_error_handler),
5214 (bus_connection_setup):
5215 * bus/main.c: (main):
5216 Make sure that the DBusConnectionData struct is NULLed
5217 out to prevent a segfault.
5219 * dbus/dbus-errors.c: (dbus_result_to_string):
5220 * dbus/dbus-errors.h:
5221 * dbus/dbus-message.c: (dbus_message_get_fields),
5222 (dbus_message_get_fields_valist), (_dbus_message_test):
5223 * dbus/dbus-message.h:
5224 Make dbus_message_get_fields return a result code so we can
5225 track invalid fields as well as oom.
5227 2003-01-11 Havoc Pennington <hp@pobox.com>
5229 * configure.in: change --enable-test/--enable-ansi action-if-given
5230 to enable_foo=$enableval instead of enable_foo=yes
5232 2003-01-08 Havoc Pennington <hp@pobox.com>
5234 * dbus/dbus-string.c (_dbus_string_align_length): new function
5236 * dbus/dbus-test-main.c: move main() for test app here
5238 (dbus_internal_symbol_do_not_use_run_tests): we have to export a
5239 symbol to run tests, because dbus-test isn't in the main
5242 Code review nitpicks.
5244 * dbus/dbus-message.c (dbus_message_write_header): add newlines
5245 for people with narrow emacs ;-). Assert client_serial was filled
5246 in. Assert message->name != NULL.
5247 (dbus_message_append_fields): have "first_field_type" arg separate
5248 from va list, needed for C++ binding that also uses varargs IIRC
5249 and helps with type safety
5250 (dbus_message_new): add @todo about using DBusString to store
5251 service/name internally
5252 (dbus_message_new): don't leak ->service and ->name on OOM later
5254 (dbus_message_unref): free the service name
5255 (dbus_message_get_fields): same change to varargs
5256 i.e. first_field_type
5257 (_dbus_message_loader_return_buffer): assert that the message data
5258 is aligned (if not it's a bug in our code). Put in verbose griping
5259 about why we set corrupted = TRUE.
5260 (decode_header_data): add FIXME that char* is evil. Was going to
5261 add FIXME about evil locale-specific string.h strncmp, but just
5262 switched to wacky string-as-uint32 optimization. Move check for
5263 "no room for field name" above get_const_data_len() to avoid
5264 assertion failure in get_const_data_len if we have trailing 2
5265 bytes or the like. Check for service and name fields being
5266 provided twice. Don't leak service/name on error. Require field
5267 names to be aligned to 4 bytes.
5269 * dbus/dbus-marshal.c: move byte swap stuff to header
5270 (_dbus_pack_int32): uscore-prefix
5271 (_dbus_unpack_int32): uscore-prefix
5272 (_dbus_unpack_uint32): export
5273 (_dbus_demarshal_string): add @todo complaining about use of
5275 (_dbus_marshal_get_field_end_pos): add @todo about bad error
5276 handling allowing corrupt data to go unchecked
5278 2003-01-08 Havoc Pennington <hp@redhat.com>
5280 * dbus/dbus-transport-unix.c (unix_do_iteration): add read/write
5281 to the select() as needed for authentication. (should be using
5282 _dbus_poll() not select, but for another day)
5284 * dbus/dbus.h: include dbus/dbus-protocol.h
5286 2003-01-08 Anders Carlsson <andersca@codefactory.se>
5288 * dbus/Makefile.am (dbusinclude_HEADERS): Install
5291 2003-01-08 Anders Carlsson <andersca@codefactory.se>
5293 * dbus/dbus-internals.c: (_dbus_type_to_string):
5294 New function that returns a string describing a type.
5296 * dbus/dbus-marshal.c: (_dbus_demarshal_byte_array):
5297 * dbus/dbus-marshal.h:
5298 * dbus/dbus-message.c: (dbus_message_get_fields_valist),
5299 (dbus_message_iter_get_field_type), (dbus_message_iter_get_double),
5300 (dbus_message_iter_get_byte_array):
5301 * dbus/dbus-message.h:
5302 Add new convenience functions for appending and getting message fields.
5303 Also add demarshalling routines for byte arrays.
5305 2003-01-07 Anders Carlsson <andersca@codefactory.se>
5307 * dbus/dbus-connection-internal.h:
5308 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
5309 (_dbus_connection_get_next_client_serial),
5310 (dbus_connection_send_message):
5311 * dbus/dbus-internals.h:
5312 * dbus/dbus-marshal.c: (unpack_uint32), (dbus_unpack_int32),
5313 (dbus_pack_int32), (_dbus_marshal_double), (_dbus_marshal_int32),
5314 (_dbus_marshal_uint32), (_dbus_demarshal_double),
5315 (_dbus_demarshal_int32), (_dbus_demarshal_uint32),
5316 (_dbus_demarshal_string), (_dbus_marshal_get_field_end_pos),
5317 (_dbus_verbose_bytes), (_dbus_marshal_test):
5318 * dbus/dbus-marshal.h:
5319 * dbus/dbus-message-internal.h:
5320 * dbus/dbus-message.c: (_dbus_message_set_client_serial),
5321 (dbus_message_write_header), (_dbus_message_lock),
5322 (dbus_message_new), (dbus_message_ref), (dbus_message_unref),
5323 (dbus_message_get_name), (dbus_message_append_int32),
5324 (dbus_message_append_uint32), (dbus_message_append_double),
5325 (dbus_message_append_string), (dbus_message_append_byte_array),
5326 (dbus_message_get_fields_iter), (dbus_message_iter_ref),
5327 (dbus_message_iter_unref), (dbus_message_iter_has_next),
5328 (dbus_message_iter_next), (dbus_message_iter_get_field_type),
5329 (dbus_message_iter_get_string), (dbus_message_iter_get_int32),
5330 (dbus_message_iter_get_uint32), (dbus_message_iter_get_double),
5331 (decode_header_data), (_dbus_message_loader_return_buffer),
5332 (message_iter_test), (_dbus_message_test):
5333 * dbus/dbus-message.h:
5334 * dbus/dbus-protocol.h:
5335 * dbus/dbus-test.c: (main):
5337 * glib/test-dbus-glib.c: (message_handler), (main):
5338 * test/echo-client.c: (main):
5339 * test/watch.c: (check_messages):
5340 Make messages sendable and receivable for real.
5342 2003-01-07 Anders Carlsson <andersca@codefactory.se>
5344 * dbus/dbus-marshal.c: (_dbus_marshal_double),
5345 (_dbus_marshal_string), (_dbus_marshal_byte_array):
5346 * dbus/dbus-message.c: (dbus_message_append_int32),
5347 (dbus_message_append_uint32), (dbus_message_append_double),
5348 (dbus_message_append_string), (dbus_message_append_byte_array):
5349 Handle OOM restoration.
5351 2003-01-07 Anders Carlsson <andersca@codefactory.se>
5353 * dbus/dbus-marshal.c: (_dbus_marshal_string),
5354 (_dbus_demarshal_string), (_dbus_marshal_test):
5355 * dbus/dbus-marshal.h:
5356 * dbus/dbus-message.c: (dbus_message_get_name),
5357 Document these functions.
5359 (dbus_message_append_int32), (dbus_message_append_uint32),
5360 (dbus_message_append_double), (dbus_message_append_string),
5361 (dbus_message_append_byte_array):
5362 * dbus/dbus-message.h:
5363 Add functions for adding message fields of different types.
5365 * dbus/dbus-protocol.h:
5366 Add the different types.
5368 2003-01-05 Havoc Pennington <hp@pobox.com>
5370 * bus/connection.c: implement routines for handling connections,
5371 first thing is keeping a list of owned services on each connection
5372 and setting up watches etc.
5374 * bus/services.c: implement a mapping from service names to lists
5377 * dbus/dbus-hash.c: add DBUS_HASH_POINTER
5379 * dbus/dbus-threads.c (dbus_static_mutex_lock): add functions
5380 to use static mutexes for global data
5382 * dbus/dbus-connection.c (dbus_connection_set_data): add new
5383 collection of functions to set/get application-specific data
5384 on the DBusConnection.
5386 2003-01-04 Havoc Pennington <hp@pobox.com>
5388 * dbus/dbus-sysdeps.c (_dbus_sleep_milliseconds): new function
5389 (_dbus_poll): new function
5391 * dbus/dbus-internals.h (_DBUS_STRUCT_OFFSET): new macro
5394 * bus/loop.c: initial code for the daemon main loop
5396 2003-01-04 Havoc Pennington <hp@pobox.com>
5398 * test/watch.c (error_handler): make it safe if the error handler
5399 is called multiple times (if we s/error handler/disconnect
5400 handler/ we should just guarantee it's called only once)
5402 * dbus/dbus-transport.c (_dbus_transport_disconnect): call the
5403 error handler on disconnect (it's quite possible we should
5404 just change the error handler to a "disconnect handler," I'm
5405 not sure we have any other meaningful errors)
5407 * configure.in: check for getpwnam_r
5409 * dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
5410 dbus/dbus-auth.c: add credentials support, add EXTERNAL auth
5411 mechanism as in SASL spec, using socket credentials
5413 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): new function
5414 (_dbus_send_credentials_unix_socket): new function
5416 * dbus/dbus-sysdeps.c (_dbus_accept_unix_socket): rename just
5418 (_dbus_write): only check errno if <0 returned
5419 (_dbus_write_two): ditto
5421 2003-01-02 Anders Carlsson <andersca@codefactory.se>
5423 * dbus/dbus-marshal.c: (_dbus_marshal_utf8_string),
5424 (_dbus_marshal_byte_array), (_dbus_demarshal_utf8_string),
5425 (_dbus_marshal_test):
5426 * dbus/dbus-marshal.h:
5427 Add _dbus_marshal_byte_array and rename _dbus_marshal_string
5428 to _dbus_marshal_utf8_string. Also fix some tests.
5430 2002-12-28 Harri Porten <porten@kde.org>
5432 * configure.in: added check for C++ compiler and a very cheesy
5433 check for the Qt integration
5435 * Makefile.am (SUBDIRS): compile qt subdir if support is enabled
5437 * qt/Makefile.am: added
5439 * qt/.cvsignore: added
5441 * qt/dbus-qthread.cc, qt/dbus-qthread.cpp: renamed former to
5442 latter, added #ifdef QT_THREAD_SUPPORT guard.
5444 * dbus/Makefile.am: added missing headers for make dist
5446 2002-12-28 Kristian Rietveld <kris@gtk.org>
5448 * dbus/Makefile.am: fixup export-symbols-regex.
5450 2002-12-27 Anders Carlsson <andersca@codefactory.se>
5452 * acinclude.m4: Add this file and put the
5453 PKG_CHECK_MODULE macro in it.
5455 2002-12-27 Anders Carlsson <andersca@codefactory.se>
5457 * dbus/dbus-marshal.c: (_dbus_marshal_string),
5458 (_dbus_demarshal_double), (_dbus_demarshal_int32),
5459 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
5460 (_dbus_marshal_test):
5461 Make the demarshalling routines align the pos argument.
5462 Add string marshalling tests and fix the obvious bugs
5465 2002-12-26 Havoc Pennington <hp@pobox.com>
5467 * dbus/dbus-auth.c: fixes fixes fixes
5469 * dbus/dbus-transport-unix.c: wire up support for
5470 encoding/decoding data on the wire
5472 * dbus/dbus-auth.c (_dbus_auth_encode_data)
5473 (_dbus_auth_decode_data): append to target string
5474 instead of nuking it.
5476 2002-12-26 Havoc Pennington <hp@pobox.com>
5478 * dbus/dbus-marshal.h (DBUS_COMPILER_BYTE_ORDER): #ifdef
5479 WORDS_BIGENDIAN then compiler byte order is DBUS_BIG_ENDIAN,
5482 * dbus/dbus-marshal.c: Add macros to do int swapping in-place and
5483 avoid swap_bytes() overhead (ignoring possible assembly stuff for
5484 now). Main point is because I wanted unpack_uint32 to implement
5486 (_dbus_verbose_bytes): new function
5488 * dbus/dbus-string.c (_dbus_string_validate_ascii): new function
5490 * dbus/dbus-message.c (_dbus_message_loader_get_is_corrupted): add
5491 mechanism to handle a corrupt message stream
5492 (_dbus_message_loader_new): fix preallocation to only prealloc,
5495 * dbus/dbus-string.c (_dbus_string_skip_blank): fix this function
5496 (_dbus_string_test): enhance tests for copy/move and fix the
5499 * dbus/dbus-transport-unix.c: Hold references in more places to
5500 avoid reentrancy problems
5502 * dbus/dbus-transport.c: ditto
5504 * dbus/dbus-connection.c (dbus_connection_dispatch_message): don't
5505 leak reference count in no-message case
5507 * test/watch.c (do_mainloop): handle adding/removing watches
5508 during iteration over the watches. Also, ref the connection/server
5509 stored on a watch, so we don't try to mangle a destroyed one.
5511 * dbus/dbus-transport-unix.c (do_authentication): perform
5514 * dbus/dbus-auth.c (get_state): add a state
5515 AUTHENTICATED_WITH_UNUSED_BYTES and return it if required
5516 (_dbus_auth_get_unused_bytes): append the unused bytes
5517 to the passed in string, rather than prepend
5519 * dbus/dbus-transport.c (_dbus_transport_init_base): create
5520 the auth conversation DBusAuth
5522 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd)
5523 (_dbus_transport_new_for_domain_socket): when creating a
5524 transport, pass in whether it's a client-side or server-side
5525 transport so we know which DBusAuth to create
5527 2002-12-03 Havoc Pennington <hp@pobox.com>
5529 * dbus/dbus-transport-unix.c (unix_finalize): finalize base
5530 _after_ finalizing the derived members
5531 (unix_connection_set): unref watch if we fail to add it
5533 * dbus/dbus-connection.c (dbus_connection_unref): delete the
5534 transport first, so that the connection owned by the
5535 transport will be valid as the transport finalizes.
5537 * dbus/dbus-transport-unix.c (unix_finalize): free the write_watch
5538 if necessary, and remove watches from the connection.
5540 * dbus/dbus-watch.c (_dbus_watch_list_free): improve a comment
5542 2002-12-26 Anders Carlsson <andersca@codefactory.se>
5544 * dbus/dbus-marshal.c: (_dbus_marshal_string),
5545 (_dbus_demarshal_double), (_dbus_demarshal_int32),
5546 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
5547 (_dbus_marshal_test):
5548 * dbus/dbus-marshal.h:
5549 Add string marshal functions and have the demarshal functions
5550 return the new position.
5552 2002-12-25 Havoc Pennington <hp@pobox.com>
5554 * doc/dbus-sasl-profile.txt: docs on the authentication protocol,
5555 it is a simple protocol that just maps directly to SASL.
5557 * dbus/dbus-auth.h, dbus/dbus-auth.c: authentication protocol
5558 initial implementation, not actually used yet.
5560 * dbus/dbus-string.c (_dbus_string_find): new function
5561 (_dbus_string_equal): new function
5562 (_dbus_string_base64_encode): new function
5563 (_dbus_string_base64_decode): new function
5565 2002-12-25 Anders Carlsson <andersca@codefactory.se>
5568 * dbus/dbus-marshal.c: (swap_bytes), (_dbus_marshal_double),
5569 (_dbus_marshal_int32), (_dbus_marshal_uint32),
5570 (_dbus_demarshal_double), (_dbus_demarshal_int32),
5571 (_dbus_demarshal_uint32), (_dbus_marshal_test):
5572 * dbus/dbus-marshal.h:
5573 * dbus/dbus-protocol.h:
5574 * dbus/dbus-test.c: (main):
5576 Add un-optimized marshalling/demarshalling routines.
5578 2002-12-25 Harri Porten <porten@kde.org>
5580 * qt/dbus-qt.h: adjusted ctor and getter to KDE/Qt conventions
5582 2002-12-24 Zack Rusin <zack@kde.org>
5584 * qt/dbus-qthread.cc: adding - integrates QMutex into Dbus
5585 * qt/dbus-qt.h: skeleton with two sample implemenatation of the
5588 2002-12-24 Havoc Pennington <hp@pobox.com>
5590 * glib/dbus-gthread.c: fix include
5592 * glib/dbus-glib.h: rename DBusMessageHandler for now.
5593 I think glib API needs to change, though, as you don't
5594 want to use DBusMessageFunction, you want to use the
5595 DBusMessageHandler object. Probably
5596 dbus_connection_open_with_g_main_loop()
5597 and dbus_connection_setup_g_main_loop() or something like that
5598 (but think of better names...) that just create a connection
5599 that has watch/timeout functions etc. already set up.
5601 * dbus/dbus-connection.c
5602 (dbus_connection_send_message_with_reply): new function just to
5603 show how the message handler helps us deal with replies.
5605 * dbus/dbus-list.c (_dbus_list_remove_last): new function
5607 * dbus/dbus-string.c (_dbus_string_test): free a string that
5610 * dbus/dbus-hash.c: use memory pools for the hash entries
5611 (rebuild_table): be more paranoid about overflow, and
5612 shrink table when we can
5613 (_dbus_hash_test): reduce number of sprintfs and write
5614 valid C89. Add tests for case where we grow and then
5615 shrink the hash table.
5617 * dbus/dbus-mempool.h, dbus/dbus-mempool.c: memory pools
5619 * dbus/dbus-connection.c (dbus_connection_register_handler)
5620 (dbus_connection_unregister_handler): new functions
5622 * dbus/dbus-message.c (dbus_message_get_name): new
5624 * dbus/dbus-list.c: fix docs typo
5626 * dbus/dbus-message-handler.h, dbus/dbus-message-handler.c:
5627 an object representing a handler for messages.
5629 2002-12-16 Anders Carlsson <andersca@codefactory.se>
5632 * glib/dbus-gthread.c: (dbus_gthread_init):
5633 Don't use the gdbus prefix for public functions.
5635 2002-12-16 Anders Carlsson <andersca@codefactory.se>
5639 Add GLib checks and fixup .pc files
5643 * glib/dbus-gmain.c: (gdbus_connection_prepare),
5644 (gdbus_connection_check), (gdbus_connection_dispatch),
5645 (gdbus_add_connection_watch), (gdbus_remove_connection_watch),
5646 (dbus_connection_gsource_new):
5647 * glib/dbus-gthread.c: (dbus_gmutex_new), (dbus_gmutex_free),
5648 (dbus_gmutex_lock), (dbus_gmutex_unlock), (dbus_gthread_init):
5649 * glib/test-dbus-glib.c: (message_handler), (main):
5652 2002-12-15 Harri Porten <porten@kde.org>
5654 * autogen.sh: check for libtoolize before attempting to use it
5656 * dbus/dbus-transport-unix.c: include <sys/time.h> for timeval
5659 * .cvsignore: ignore more stamp files
5661 * dbus/dbus-watch.c (_dbus_watch_list_new): fixed doc error
5663 * test/Makefile.am: added -I$(top_srcdir) to be able to compile
5664 without make install.
5666 2002-12-15 Havoc Pennington <hp@pobox.com>
5668 * dbus/dbus-threads.c: add thread stubs that a higher library
5669 layer can fill in. e.g. the GLib wrapper might fill them in with
5670 GThread stuff. We still need to use this thread API to
5671 thread-safe-ize the library.
5673 2002-12-12 Havoc Pennington <hp@pobox.com>
5675 * dbus/dbus-transport-unix.c, dbus/dbus-server-unix.c: use the
5676 below new interfaces and include fewer system headers.
5678 * dbus/dbus-sysdeps.c (_dbus_read): new function
5679 (_dbus_write): new function
5680 (_dbus_write_two): new function
5681 (_dbus_connect_unix_socket): new function
5682 (_dbus_listen_unix_socket): new function
5684 * dbus/dbus-message-internal.h: change interfaces to use
5687 2002-12-11 Havoc Pennington <hp@pobox.com>
5689 * dbus/dbus-types.h: add dbus_unichar
5691 * dbus/dbus-internals.c (_dbus_verbose): use _dbus_getenv
5693 * dbus/dbus-connection.c (dbus_connection_send_message): return
5696 * dbus/dbus-transport.c: include dbus-watch.h
5698 * dbus/dbus-connection.c: include dbus-message-internal.h
5700 * HACKING: add file with coding guidelines stuff.
5702 * dbus/dbus-string.h, dbus/dbus-string.c: Encapsulate all string
5703 handling here, for security purposes (as in vsftpd). Not actually
5704 using this class yet.
5706 * dbus/dbus-sysdeps.h, dbus/dbus-sysdeps.c: Encapsulate all
5707 system/libc usage here, as in vsftpd, for ease of auditing (and
5708 should also simplify portability). Haven't actually moved all the
5709 system/libc usage into here yet.
5711 2002-11-25 Havoc Pennington <hp@pobox.com>
5713 * dbus/dbus-internals.c (_dbus_verbose): fix to not
5714 always print the first verbose message.
5716 2002-11-24 Havoc Pennington <hp@pobox.com>
5718 * test/echo-client.c, test/echo-server.c: cheesy test
5721 * configure.in (AC_CHECK_FUNCS): check for writev
5723 * dbus/dbus-message.c (_dbus_message_get_network_data): new
5726 * dbus/dbus-list.c (_dbus_list_foreach): new function
5728 * dbus/dbus-internals.c (_dbus_verbose): new function
5730 * dbus/dbus-server.c, dbus/dbus-server.h: public object
5731 representing a server that listens for connections.
5733 * dbus/.cvsignore: create
5735 * dbus/dbus-errors.h, dbus/dbus-errors.c:
5736 public API for reporting errors
5738 * dbus/dbus-connection.h, dbus/dbus-connection.c:
5739 public object representing a connection that
5740 sends/receives messages. (Same object used for
5741 both client and server.)
5743 * dbus/dbus-transport.h, dbus/dbus-transport.c:
5744 Basic abstraction for different kinds of stream
5745 that we might read/write messages from.
5747 2002-11-23 Havoc Pennington <hp@pobox.com>
5749 * dbus/dbus-internals.h (_DBUS_INT_MAX): add _DBUS_INT_MIN
5752 * dbus/dbus-test.c (main): add list test, and include
5753 dbus-test.h as intended
5755 * dbus/dbus-hash.c (_dbus_hash_table_remove_string)
5756 (_dbus_hash_table_remove_int): return value indicates
5757 whether the entry existed to remove
5759 * dbus/dbus-list.c: add linked list utility class,
5762 * dbus/dbus-hash.c: add TODO item about shrinking the hash bucket
5765 2002-11-23 Havoc Pennington <hp@pobox.com>
5767 * Doxyfile.in (INCLUDE_FILE_PATTERNS): expand DBUS_BEGIN_DECLS/
5768 DBUS_END_DECLS to nothing, that should fix this once and for all
5770 * Doxyfile.in (JAVADOC_AUTOBRIEF): set to YES
5772 * dbus/dbus-message.c, dbus/dbus-hash.c:
5773 add some missing @brief
5775 2002-11-23 Havoc Pennington <hp@pobox.com>
5777 * dbus/dbus-message.h: put semicolons after DEBUG_BEGIN_DECLS
5778 to avoid confusing Doxygen
5780 * dbus/dbus-hash.c: @} not }@
5782 * dbus/dbus-message.c (struct DBusMessage): split out
5785 2002-11-23 Havoc Pennington <hp@pobox.com>
5787 * configure.in: pile on more warning flags if using gcc
5789 * Doxyfile.in (EXTRACT_STATIC): set to NO, so we don't have
5790 to document static functions
5792 * configure.in: add summary to end of configure so it
5793 looks nice and attractive
5795 * dbus/dbus-hash.c: finish implementation and write unit
5798 * configure.in: add --enable-tests to enable unit tests
5800 * dbus/dbus-test.c: test program to run unit tests
5801 for all files in dbus/*, initially runs a test for
5804 * dbus/dbus-internals.h: file to hold some internal utility stuff
5806 2002-11-22 Havoc Pennington <hp@redhat.com>
5808 * dbus/dbus-hash.c: copy in Tcl hash table, not yet
5809 "ported" away from Tcl
5811 * dbus/dbus-types.h: header for types such as dbus_bool_t
5813 2002-11-22 Havoc Pennington <hp@redhat.com>
5815 * dbus/dbus.h: fixups for doc warnings
5817 * Doxyfile.in (FILE_PATTERNS): we need to scan .h to pick up
5819 (QUIET): make it quiet so we can see warnings
5821 * dbus/dbus-memory.c: teach D-BUS to allocate and free memory
5823 2002-11-22 Havoc Pennington <hp@redhat.com>
5825 * Makefile.am: include "Doxyfile" target in all-local
5827 * configure.in: generate the Doxyfile
5829 * Doxyfile.in: move Doxyfile here, so we can use
5830 configure to generate a Doxyfile with the right
5833 2002-11-22 Havoc Pennington <hp@redhat.com>
5835 * dbus/dbus-message.c: move inline docs into .c file
5837 * Doxyfile (OUTPUT_DIRECTORY): move output to doc/api
5838 so all docs are under doc/
5839 (MAN_EXTENSION): generate man pages. Use extension
5840 ".3dbus" which matches ".3qt" on my system,
5841 I guess this is OK, I don't know really.
5842 (FILE_PATTERNS): look for .c files not .h, makes sense
5845 2002-11-22 Havoc Pennington <hp@pobox.com>
5847 * Makefile.am (SUBDIRS): rename subdir "server" to "bus"
5848 because any app can be a server, and any app can be a client,
5849 the bus is a special kind of server.
5851 Thu Nov 21 23:35:31 2002 Zack Rusin <zack@kde.org>
5853 * Doxyfile : adding. Still needs Makefile rules to be generated
5854 automatically (just run "doxygen" in the toplevel dir for now to
5857 * dbus/dbus-message.h : Adding sample docs (javadoc since
5858 resembles gtk-doc a little more)
5860 * dbus/dbus.h : Adding sample docs
5862 2002-11-21 Havoc Pennington <hp@redhat.com>
5864 * dbus/Makefile.am (INCLUDES): define DBUS_COMPILATION
5865 so we can allow ourselves to include files directly,
5866 instead of having to use dbus.h
5868 * dbus/dbus.h: fill in
5870 * dbus/dbus-message.h: sketch out a sample header file.
5871 Include griping if you include it directly instead of
5874 * dbus/dbus-macros.h: new file with macros for extern "C",
5875 TRUE/FALSE, NULL, etc.
5877 * doc/file-boilerplate.c: put include guards in here
5879 2002-11-21 Havoc Pennington <hp@redhat.com>
5881 * doc/file-boilerplate.c: include both AFL and GPL boilerplate.
5883 * COPYING: include the GPL as well, and license code
5884 under both AFL and GPL.
5886 2002-11-21 Havoc Pennington <hp@redhat.com>
5888 * acconfig.h: get rid of this
5890 * autogen.sh (run_configure): add --no-configure option
5892 * configure.in: remove AC_ARG_PROGRAM to make
5893 autoconf complain less. add AC_PREREQ.
5894 add AC_DEFINE third arg.
5896 2002-11-21 Anders Carlsson <andersca@codefactory.se>
5899 Fix references so we can distcheck.
5901 2002-11-21 Havoc Pennington <hp@redhat.com>
5903 * Initial module creation