1 2004-03-27 Havoc Pennington <hp@redhat.com>
3 Patch from Timo Teräs:
5 * tools/dbus-send.c (main): if --print-reply, assume type is
6 method call; support boolean type args
8 * dbus/dbus-connection.c (dbus_connection_send_with_reply): fix a
9 bunch of memleak and logic bugs
11 2004-03-23 Owen Fraser-Green <owen@discobabe.net>
14 * mono/Introspector.cs:
16 * mono/InterfaceProxy.cs:
18 * mono/ProxyBuilder.cs:
20 Added InterfaceProxy class to avoid building proxies for every
24 * dbus-message.c (dbus_message_append_args_valist)
25 (dbus_message_iter_get_object_path)
26 (dbus_message_iter_get_object_path_array)
27 (dbus_message_iter_append_object_path)
28 (dbus_message_iter_append_object_path_array):
29 Added object_path iter functions to handle OBJECT_PATH arguments
31 2004-03-23 Owen Fraser-Green <owen@discobabe.net>
33 First checkin of mono bindings.
36 Build stuff for the bindings
37 * dbus-sharp.pc.in: Added for pkgconfig
39 2004-03-21 Havoc Pennington <hp@redhat.com>
41 * test/test-service.c (main): remove debug spew
43 2004-03-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
45 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): accept empty
48 * dbus/dbus-message.h, bus/dbus-message.c (dbus_message_iter_init)
49 (dbus_message_iter_init_array_iterator)
50 (dbus_message_iter_init_dict_iterator): return a dbus_bool_t to
51 indicate whether the iterator is empty
53 * dbus/dbus-pending-call.c, dbus/dbus-server.c: silence compiler
56 2004-03-19 Havoc Pennington <hp@redhat.com>
62 * doc/Makefile.am: add all XMLTO usage to DBUS_XML_DOCS_ENABLED
64 * python/Makefile.am: change to avoid dist of dbus_bindings.c so
65 you don't need pyrex to make dist
67 * qt/Makefile.am (libdbus_qt_1_la_SOURCES): add integrator.h to
70 2004-03-18 Richard Hult <richard@imendio.com>
72 * dbus/dbus-message.c (dbus_message_get_auto_activation)
73 (dbus_message_set_auto_activation): Add doxygen docs.
75 2004-03-16 Richard Hult <richard@imendio.com>
77 * bus/activation.c: (bus_activation_service_created),
78 (bus_activation_send_pending_auto_activation_messages),
79 (bus_activation_activate_service):
81 * bus/dispatch.c: (bus_dispatch),
82 (check_nonexistent_service_auto_activation),
83 (check_service_auto_activated),
84 (check_segfault_service_auto_activation),
85 (check_existent_service_auto_activation), (bus_dispatch_test):
86 * bus/driver.c: (bus_driver_handle_activate_service):
87 * bus/services.c: (bus_registry_acquire_service):
88 * dbus/dbus-message.c: (dbus_message_set_auto_activation),
89 (dbus_message_get_auto_activation):
90 * dbus/dbus-message.h:
91 * dbus/dbus-protocol.h: Implement auto-activation.
93 * doc/dbus-specification.xml: Add auto-activation to the spec.
95 2004-03-12 Olivier Andrieu <oliv__a@users.sourceforge.net>
97 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos):
98 fix a bug with CUSTOM types.
100 * dbus/dbus-message.c (message_iter_test, _dbus_message_test): add
101 a unit test for this bug (used to fail).
103 2004-03-12 Mikael Hallendal <micke@imendio.com>
106 (babysitter_watch_callback): notify all pending activations waiting for
107 the same exec that the activation failed.
108 (bus_activation_activate_service): shortcut the activation if we
109 already waiting for the same executable to start up.
111 2004-03-12 Mikael Hallendal <micke@imendio.com>
114 - Added service file reloading.
115 Each service files directory is kept in an hash table in
116 BusActivation and each BusActivationEntry knows what .service-file it
117 was read from. So when you try to activate a service the bus will
118 check if it's been updated, removed or if new .service-files has
120 - Test code at the bottom for the service file reloading.
121 * bus/test-main.c: (main):
123 - added service reloading test.
124 * dbus/dbus-sysdeps.c:
125 * dbus/dbus-sysdeps.h: (_dbus_delete_directory): Added.
127 2004-03-08 Michael Meeks <michael@ximian.com>
129 * dbus/dbus-connection.c (_dbus_connection_block_for_reply):
130 bail immediately if disconnected, to avoid busy loop.
132 * dbus/dbus-message.c (dbus_message_iter_get_args_valist):
133 cleanup cut/paste/inefficiency.
135 2004-03-01 David Zeuthen <david@fubar.dk>
137 * dbus/dbus-string.c (_dbus_string_append_printf_valist): Fix a
138 bug where args were used twice. This bug resulted in a segfault
139 on a Debian/PPC system when starting the messagebus daemon. Include
140 dbus-sysdeps.h for DBUS_VA_COPY
142 * dbus/dbus-sysdeps.h: Define DBUS_VA_COPY if neccessary. From GLib
144 * configure.in: Check for va_copy; define DBUS_VA_COPY to the
145 appropriate va_copy implementation. From GLib
147 2004-02-24 Joe Shaw <joe@ximian.com>
149 * bus/services.c (bus_registry_acquire_service): We need to pass
150 in the service name to dbus_set_error() to prevent a crash.
152 2003-12-26 Anders Carlsson <andersca@gnome.org>
154 * AUTHORS: Reveal my True identity.
156 2003-12-17 Mikael Hallendal <micke@imendio.com>
158 * dbus/dbus-message.c: (dbus_message_append_args_valist):
159 - Added case for DBUS_TYPE_BYTE, patch from Johan Hedberg.
161 2003-12-13 Mikael Hallendal <micke@imendio.com>
163 * doc/TODO: Added not about better error check of configuration files.
165 2003-12-02 Richard Hult <richard@imendio.com>
167 * Update AFL version to 2.0 throughout the source files to reflect
168 the update that was done a while ago.
170 2003-12-02 Richard Hult <richard@imendio.com>
172 * dbus/dbus-message.c (dbus_message_iter_append_dict): Set
173 wrote_dict_key to FALSE on the iter that the dict is appended to,
174 just like when appending other types. Fixes a bug where a dict
175 couldn't be put inside a dict.
176 (dbus_message_iter_append_dict_key): Fix typo in warning message.
177 (message_iter_test, _dbus_message_test): Add test case for dict
180 2003-12-01 David Zeuthen <david@fubar.dk>
182 * python/dbus.py: Add the actual message when calling the reciever
183 of a signal such that parameters can be inspected. Add the method
184 remove_signal_receiver
186 2003-11-26 Mikael Hallendal <micke@imendio.com>
190 * glib/*.[ch]: Made ref functions return the pointer
192 2003-11-25 Zack Rusin <zack@kde.org>
194 * qt/integrator.h, qt/integrator.cpp: Adding handling of DBusServer,
196 * qt/server.h, qt/server.cpp, qt/Makefile.am: Adding DBusServer
199 * qt/connection.h, qt/connection.cpp: Adjusting to changes in
200 the Integrator and to better fit with the server,
202 2003-11-24 Zack Rusin <zack@kde.org>
204 * qt/connection.h, qt/connection.cpp: removing initDbus method since
205 the integrator handles it now
207 * qt/integrator.h, qt/integrator.cpp: reworking handling of timeouts,
208 since QTimer wasn't really meant to be used the way DBusTimeout is
210 2003-11-24 Zack Rusin <zack@kde.org>
212 * qt/integrator.h, qt/integrator.cpp, Makefile.am: Adding
213 Integrator class which integrates D-BUS with the Qt event loop,
215 * qt/connection.h, qt/connection.cpp: Move all the code which
216 was dealing with D-BUS integration to the Integrator class,
217 and start using Integrator,
219 2003-11-23 Zack Rusin <zack@kde.org>
221 * qt/connection.h, qt/connection.cpp: Adding the DBusConnection
224 * qt/message.h, qt/message.cpp: updating to the current D-BUS api,
225 switching namespaces to DBusQt, reworking the class,
227 * Makefile.cvs: switching dependencies so that it matches KDE
230 * qt/Makefile.am: adding connection.{h,cpp} and message.{h,cpp} to
233 2003-11-19 Havoc Pennington <hp@redhat.com>
237 * configure.in: bump version to 0.20
239 * configure.in (have_qt): add yet another place to look for qt
240 (someone hand trolltech a .pc file...)
242 2003-11-01 Havoc Pennington <hp@redhat.com>
244 * doc/dbus-specification.xml: add state machine docs on the auth
245 protocol; just a first draft, I'm sure it's wrong.
247 2003-10-28 David Zeuthen <david@fubar.dk>
249 * python/dbus_bindings.pyx.in: add get_dict to handle dictionaries
250 return types. Fixup TYPE_* to reflect changes in dbus/dbus-protocol.h
252 2003-10-28 Havoc Pennington <hp@redhat.com>
254 * dbus/dbus-message.c (get_next_field): delete unused function
256 2003-10-28 Havoc Pennington <hp@redhat.com>
258 * bus/expirelist.c (do_expiration_with_current_time): detect
259 failure of the expire_func due to OOM
261 * bus/connection.c (bus_pending_reply_expired): return FALSE on OOM
263 * bus/dispatch.c (check_send_exit_to_service): fix to handle the
264 NoReply error that's now created by the bus when the service exits
266 2003-10-28 Havoc Pennington <hp@redhat.com>
268 * dbus/dbus-message.c (_dbus_message_test): enable and fix the
269 tests for set_path, set_interface, set_member, etc.
271 * dbus/dbus-string.c (_dbus_string_insert_bytes): allow 0 bytes
273 * dbus/dbus-message.c (set_string_field): always just delete and
274 re-append the field; accept NULL for deletion
275 (re_align_fields_recurse): reimplement
277 2003-10-26 Havoc Pennington <hp@redhat.com>
279 * dbus/dbus-connection.c: fix docs to properly describe the
281 (_dbus_connection_notify_disconnected): remove this function;
282 we can't synchronously add the disconnected message, we have to
283 do it after we've queued any remaining real messages
284 (_dbus_connection_get_dispatch_status_unlocked): queue the
285 disconnect message only if the transport has finished queueing all
286 its real messages and is disconnected.
287 (dbus_connection_disconnect): update the dispatch status here
289 2003-10-22 Havoc Pennington <hp@redhat.com>
291 * bus/bus.c (bus_context_check_security_policy): fix up assertion
293 * bus/connection.c (bus_transaction_send_from_driver): set the
294 destination to the connection's base service
296 2003-10-20 Havoc Pennington <hp@redhat.com>
298 hmm, make check is currently not passing.
300 * doc/dbus-specification.xml: add requirement that custom type
301 names follow the same rules as interface names.
303 * dbus/dbus-protocol.h: change some of the byte codes, to avoid
304 duplication and allow 'c' to be 'custom'; dict is now 'm' for
307 * doc/dbus-specification.xml: update type codes to match
308 dbus-protocol.h, using the ASCII byte values. Rename type NAMED to
309 CUSTOM. Add type OBJECT_PATH to the spec.
311 2003-10-17 Havoc Pennington <hp@redhat.com>
313 * bus/driver.c (create_unique_client_name): use "." as separator
314 in base service names instead of '-'
316 * dbus/dbus-string.c (_dbus_string_get_byte): allow getting nul
317 byte at the end of the string
319 * dbus/dbus-internals.h (_DBUS_LIKELY, _DBUS_UNLIKELY): add
320 optimization macros since string validation seems to be a slow
323 * doc/dbus-specification.xml: restrict valid
324 service/interface/member/error names. Add test suite code for the
327 * dbus/dbus-string.c: limit service/interface/member/error names
330 * dbus/dbus-connection.c (dbus_connection_dispatch): add missing
331 format arg to verbose spew
333 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): if not out of
334 memory, return instead of g_error
336 * test/test-service.c (path_message_func): support emitting a
339 * dbus/dbus-bus.c (init_connections_unlocked): only fill in
340 activation bus type if DBUS_BUS_ACTIVATION was set; default to
341 assuming the activation bus was the session bus so that services
342 started manually will still register.
343 (init_connections_unlocked): fix so that in OOM situation we get
344 the same semantics when retrying the function
346 * test/test-service.c (main): change to use path registration, to
347 test those codepaths; register with DBUS_BUS_ACTIVATION rather
348 than DBUS_BUS_SESSION
350 2003-10-16 Havoc Pennington <hp@redhat.com>
352 * glib/dbus-gtest-main.c: bracket with #ifdef DBUS_BUILD_TESTS
354 * Makefile.am (GCOV_DIRS): remove "test", we don't care about test
355 coverage of the tests
356 (coverage-report.txt): don't move the .da and .bbg files around
358 2003-10-16 Havoc Pennington <hp@redhat.com>
360 * bus/bus.c (struct BusContext): remove struct field I didn't mean
363 2003-10-16 Havoc Pennington <hp@redhat.com>
365 * bus/connection.c (bus_pending_reply_expired): either cancel or
367 (bus_connections_check_reply): use unlink, not remove_link, as we
368 don't want to free the link; fixes double free mess
370 * dbus/dbus-pending-call.c (dbus_pending_call_block): fix in case
371 where no reply was received
373 * dbus/dbus-connection.c (_dbus_pending_call_complete_and_unlock):
376 * bus/signals.c (match_rule_matches): add special cases for the
377 bus driver, so you can match on sender/destination for it.
379 * dbus/dbus-sysdeps.c (_dbus_abort): print backtrace if
380 DBUS_PRINT_BACKTRACE is set
382 * dbus/dbus-internals.c: add pid to assertion failure messages
384 * dbus/dbus-connection.c: add message type code to the debug spew
386 * glib/dbus-gproxy.c (gproxy_get_match_rule): match rules want
387 sender=foo not service=foo
389 * dbus/dbus-bus.c (dbus_bus_get): if the activation bus is the
390 session bus but DBUS_SESSION_BUS_ADDRESS isn't set, use
391 DBUS_ACTIVATION_ADDRESS instead
393 * bus/activation.c: set DBUS_SESSION_BUS_ADDRESS,
394 DBUS_SYSTEM_BUS_ADDRESS if appropriate
396 * bus/bus.c (bus_context_new): handle OOM copying bus type into
399 * dbus/dbus-message.c (dbus_message_iter_get_object_path): new function
400 (dbus_message_iter_get_object_path_array): new function (half
401 finished, disabled for the moment)
403 * glib/dbus-gproxy.c (dbus_gproxy_end_call): properly handle
404 DBUS_MESSAGE_TYPE_ERROR
406 * tools/dbus-launch.c (babysit): support DBUS_DEBUG_OUTPUT to
407 avoid redirecting stderr to /dev/null
408 (babysit): close stdin if not doing the "exit_with_session" thing
410 * dbus/dbus-sysdeps.c (_dbus_become_daemon): delete some leftover
411 debug code; change DBUS_DEBUG_OUTPUT to only enable stderr, not
412 stdout/stdin, so things don't get confused
414 * bus/system.conf.in: fix to allow replies, I modified .conf
415 instead of .conf.in again.
417 2003-10-14 David Zeuthen <david@fubar.dk>
419 * python/dbus_bindings.pyx.in (MessageIter.get): fixed typo in
420 argtype to arg_type when raising unknown arg type exception.
421 Changed type list to reflect the changes in dbus-protocol.h so
422 the bindings actually work.
424 2003-10-14 Havoc Pennington <hp@redhat.com>
426 * test/decode-gcov.c: support gcc 3.3 also, though gcc 3.3 seems
427 to have a bug keeping it from outputting the .da files sometimes
428 (string_get_string): don't append garbage nul bytes to the string.
430 2003-10-15 Seth Nickell <seth@gnome.org>
432 * python/Makefile.am:
434 Include dbus_h_wrapper.h in the dist tarball.
436 2003-10-14 Havoc Pennington <hp@redhat.com>
438 * bus/bus.c (bus_context_check_security_policy): revamp this to
439 work more sanely with new policy-based requested reply setup
441 * bus/connection.c (bus_transaction_send_from_driver): set bus
442 driver messages as no reply
444 * bus/policy.c (bus_client_policy_check_can_receive): handle a
445 requested_reply attribute on allow/deny rules
447 * bus/system.conf: add <allow requested_reply="true"/>
449 * bus/driver.c (bus_driver_handle_message): fix check for replies
450 sent to the bus driver, which was backward. How did this ever work
451 at all though? I think I'm missing something.
453 * dbus/dbus-message.c (decode_header_data): require error and
454 method return messages to have a reply serial field to be valid
455 (_dbus_message_loader_queue_messages): break up this function;
456 validate that reply serial and plain serial are nonzero;
457 clean up the OOM/error handling.
458 (get_uint_field): don't return -1 from this
459 (dbus_message_create_header): fix signed/unsigned bug
461 * bus/connection.c (bus_connections_expect_reply): save serial of
462 the incoming message, not reply serial
464 2003-10-14 Havoc Pennington <hp@redhat.com>
466 * bus/connection.c: implement pending reply tracking using
469 * bus/bus.c (bus_context_check_security_policy): verify that a
470 reply is pending in order to allow a reply to be sent. Deny
471 messages of unknown type.
473 * bus/dbus-daemon-1.1.in: update to mention new resource limits
475 * bus/bus.c (bus_context_get_max_replies_per_connection): new
476 (bus_context_get_reply_timeout): new
478 2003-10-13 Seth Nickell <seth@gnome.org>
480 * python/Makefile.am:
482 Pass "make distcheck": remove a couple files from DIST_FILES
483 that weren't included in the final version.
485 2003-10-12 Havoc Pennington <hp@pobox.com>
487 Added test code that 1) starts an actual bus daemon and 2) uses
488 DBusGProxy; fixed bugs that were revealed by the test. Lots
489 more testing possible, but this is the basic framework.
491 * glib/dbus-gproxy.c (dbus_gproxy_manager_unregister): remove
492 empty proxy lists from the proxy list hash
494 * dbus/dbus-message.c (dbus_message_iter_get_args_valist): add a
495 couple of return_if_fail checks
497 * dbus/dbus-pending-call.c (_dbus_pending_call_new): use dbus_new0
498 to allocate, so everything is cleared to NULL as it should be.
500 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): pass
501 source as data to dbus_connection_set_timeout_functions() as the
502 timeout functions expected
504 * test/glib/run-test.sh: add a little script to start up a message
505 bus and run tests using it
507 * tools/dbus-launch.1: updates
509 * tools/dbus-launch.c (main): add --config-file option
511 * tools/dbus-launch.c (main): remove confusing else if (runprog)
512 that could never be reached.
514 * dbus/dbus-message.c (dbus_message_new_method_return)
515 (dbus_message_new_error, dbus_message_new_signal): set the
516 no-reply-expected flag on all these. Redundant, but may
517 as well be consistent.
519 2003-10-11 Havoc Pennington <hp@pobox.com>
521 * test/decode-gcov.c (function_solve_graph): make broken block
522 graph a nonfatal error since it seems to be broken. Need to debug
525 * dbus/dbus-marshal.c (_dbus_type_is_valid): new function since we
526 can't just check type > INVALID < LAST anymore
528 * dbus/dbus-message.c (dbus_message_get_signature): new function
529 (dbus_message_has_signature): new function
530 (struct DBusMessage): add signature field (right now it isn't sent
531 over the wire, just generated on the fly)
532 (dbus_message_copy): copy the signature, and init strings to
533 proper length to avoid some reallocs
534 (dbus_message_iter_init_array_iterator): return void, since it
536 (dbus_message_iter_init_dict_iterator): return void since it can't fail
537 (_dbus_message_loader_queue_messages): add silly temporary hack to
538 fill in message->signature on load
540 * dbus/dbus-protocol.h: change DBUS_TYPE_* values to be ASCII
541 characters, so they are relatively human-readable.
543 2003-10-11 Havoc Pennington <hp@pobox.com>
545 * dbus/dbus-message.c (_dbus_message_test): add more test
546 coverage, but #if 0 for now since they uncover a bug
547 not fixed yet; I think in re_align_field_recurse()
548 (re_align_field_recurse): add FIXME about broken assertion
550 * dbus/dbus-sysdeps.c (_dbus_sysdeps_test): add more test coverage
552 * bus/connection.c: share a couple code bits with expirelist.c
554 * bus/expirelist.h, bus/expirelist.c: implement a generic
555 expire-items-after-N-seconds facility, was going to share between
556 expiring connections and replies, decided not to use for expiring
559 * COPYING: include AFL 2.0 (still need to change all the file headers)
561 2003-10-09 Havoc Pennington <hp@redhat.com>
563 * configure.in: define DBUS_HAVE_GCC33_GCOV if we have
564 gcc 3.3. Not that we do anything about it yet.
566 * bus/signals.c (bus_match_rule_parse): impose max length on the
569 * dbus/dbus-protocol.h: add DBUS_MAXIMUM_MATCH_RULE_LENGTH
571 2003-10-09 Havoc Pennington <hp@redhat.com>
573 Make matching rules theoretically work (add parser).
575 * bus/bus.c (bus_context_check_security_policy): fix up to handle
576 the case where destination is explicitly specified as bus driver
577 and someone else is eavesdropping.
579 * bus/policy.c (bus_client_policy_check_can_receive): fix up
580 definition of eavesdropping and assertion
582 * tools/dbus-send.c (main): use dbus_message_type_from_string
584 * bus/signals.c (bus_match_rule_parse): implement
586 * dbus/dbus-message.c (dbus_message_type_from_string): new
588 * dbus/dbus-errors.h (DBUS_ERROR_MATCH_RULE_INVALID): add
590 2003-10-02 Havoc Pennington <hp@pobox.com>
592 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): rename from
593 dbus_gproxy_oneway_call
595 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main)
596 (dbus_server_setup_with_g_main): fix to allow calling them more
597 than once on the same args
598 (dbus_bus_get_with_g_main): new function
600 2003-10-02 Havoc Pennington <hp@redhat.com>
602 * doc/dbus-tutorial.xml: write some stuff
604 2003-09-29 Havoc Pennington <hp@pobox.com>
606 * configure.in: split checks for Doxygen from XML docs, check for
609 * doc/Makefile.am: XML-ify all the docs, and add a blank
612 2003-09-29 Havoc Pennington <hp@pobox.com>
614 * Merge dbus-object-names branch. To see the entire patch
615 do cvs diff -r DBUS_OBJECT_NAMES_BRANCHPOINT -r dbus-object-names,
617 To revert, I tagged DBUS_BEFORE_OBJECT_NAMES_MERGE.
619 2003-09-28 Havoc Pennington <hp@pobox.com>
621 * HACKING: update to reflect new server
623 2003-09-26 Seth Nickell <seth@gnome.org>
626 * python/examples/example-signals.py:
628 Start implementing some notions of signals. The API
629 is really terrible, but they sort of work (with the
630 exception of being able to filter by service, and to
631 transmit signals *as* a particular service). Need to
632 figure out how to make messages come from the service
635 * python/dbus_bindings.pyx.in:
637 Removed duplicate message_handler callbacks.
639 2003-09-25 Havoc Pennington <hp@redhat.com>
641 * bus/session.conf.in: fix my mess
643 2003-09-25 Havoc Pennington <hp@pobox.com>
645 * bus/session.conf.in: fix security policy, reported by Seth Nickell
647 2003-09-25 Seth Nickell <seth@gnome.org>
649 * python/examples/example-service.py:
651 Johan notices complete wrong code in example-service, but
652 completely wrong in a way that works exactly the same (!).
653 Johan is confused, how could this possibly work? Example
654 code fails to serve purpose of making things clear.
657 2003-09-25 Mark McLoughlin <mark@skynet.ie>
659 * doc/dbus-specification.sgml: don't require header fields
660 to be 4-byte aligned and specify that fields should be
661 distinguished from padding by the fact that zero is not
664 * doc/TODO: remove re-alignment item and add item to doc
665 the OBJECT_PATH type.
667 * dbus/dbus-message.c:
668 (HeaderField): rename the original member to value_offset
669 and introduce a name_offset member to keep track of where
670 the field actually begins.
671 (adjust_field_offsets): remove.
672 (append_int_field), (append_uint_field),
673 (append_string_field): don't align the start of the header
674 field to a 4-byte boundary.
675 (get_next_field): impl finding the next marhsalled field
677 (re_align_field_recurse): impl re-aligning a number of
678 already marshalled fields.
679 (delete_field): impl deleting a field of any type and
680 re-aligning any following fields.
681 (delete_int_or_uint_field), (delete_string_field): remove.
682 (set_int_field), (set_uint_field): no need to re-check
683 that we have the correct type for the field.
684 (set_string_field): ditto and impl re-aligning any
686 (decode_header_data): update to take into account that
687 the fields aren't 4-byte aligned any more and the new
688 way to distinguish padding from header fields. Also,
689 don't exit when there is too much header padding.
690 (process_test_subdir): print the directory.
691 (_dbus_message_test): add test to make sure a following
692 field is re-aligned correctly after field deletion.
694 * dbus/dbus-string.[ch]:
695 (_dbus_string_insert_bytes): rename from insert_byte and
696 allow the insert of multiple bytes.
697 (_dbus_string_test): test inserting multiple bytes.
699 * dbus/dbus-marshal.c: (_dbus_marshal_set_string): add
700 warning note to docs about having to re-align any
701 marshalled values following the string.
703 * dbus/dbus-message-builder.c:
704 (append_string_field), (_dbus_message_data_load):
705 don't align the header field.
707 * dbus/dbus-auth.c: (process_test_subdir): print the
710 * test/break-loader.c: (randomly_add_one_byte): upd. for
713 * test/data/invalid-messages/bad-header-field-alignment.message:
716 * test/data/valid-messages/unknown-header-field.message: shove
717 a dict in the unknown field.
719 2003-09-25 Seth Nickell <seth@gnome.org>
722 * python/dbus_bindings.pyx.in:
724 Handle return values.
726 * python/examples/example-client.py:
727 * python/examples/example-service.py:
729 Pass back return values from the service to the client.
731 2003-09-24 Seth Nickell <seth@gnome.org>
735 Connect Object methods (when you are sharing an object) up... pass
736 in a list of methods to be shared. Sharing all the methods just
737 worked out too weird. You can now create nice Services over the
740 * python/dbus_bindings.pyx.in:
742 Keep references to user_data tuples passed into C functions so
743 Python doesn't garbage collect on us.
745 Implement MethodReturn and Error subclasses of Message for creating
746 DBusMessage's of those types.
748 * python/examples/example-client.py:
749 * python/examples/example-service.py:
751 Simple example code showing both how create DBus services and objects,
754 2003-09-23 Havoc Pennington <hp@pobox.com>
756 * glib/dbus-gproxy.c (dbus_gproxy_manager_filter): implement
758 2003-09-23 Havoc Pennington <hp@redhat.com>
760 * glib/dbus-gproxy.c (dbus_gproxy_connect_signal): implement
761 (dbus_gproxy_disconnect_signal): implement
762 (dbus_gproxy_manager_remove_signal_match): implement
763 (dbus_gproxy_manager_add_signal_match): implement
764 (dbus_gproxy_oneway_call): implement
766 2003-09-23 Havoc Pennington <hp@pobox.com>
768 * glib/dbus-gproxy.c (struct DBusGProxy): convert to a GObject
769 subclass. This means dropping the transparent thread safety of the
770 proxy; you now need a separate proxy per-thread, or your own
771 locking on the proxy. Probably right anyway.
772 (dbus_gproxy_ref, dbus_gproxy_unref): nuke, just use g_object_ref
774 2003-09-22 Havoc Pennington <hp@redhat.com>
776 * glib/dbus-gproxy.c (dbus_gproxy_manager_get): implement
778 2003-09-21 Seth Nickell <seth@gnome.org>
780 First checkin of the Python bindings.
783 * python/Makefile.am:
784 * python/dbus_bindings.pyx.in:
785 * python/dbus_h_wrapper.h:
787 Pieces for Pyrex to operate on, building a dbus_bindings.so
788 python module for low-level access to the DBus APIs.
792 High-level Python module for accessing DBus objects.
797 Build stuff for the python bindings.
801 Extra macro needed for finding the Python C header files.
803 2003-09-21 Havoc Pennington <hp@pobox.com>
805 * glib/dbus-gproxy.c (dbus_gproxy_manager_new): start
806 implementing the proxy manager, didn't get very far.
808 * dbus/dbus-bus.c (dbus_bus_add_match): new
809 (dbus_bus_remove_match): new
811 * glib/dbus-gproxy.c (dbus_gproxy_new_for_service): add a
812 path_name argument; adjust the other not-yet-implemented
813 gproxy constructors to be what I think they should be.
815 2003-09-21 Havoc Pennington <hp@pobox.com>
817 * dbus/dbus-bus.c (dbus_bus_get): set exit_on_disconnect to TRUE
818 by default for message bus connections.
820 * dbus/dbus-connection.c (dbus_connection_dispatch): exit if
821 exit_on_disconnect flag is set and we process the disconnected
823 (dbus_connection_set_exit_on_disconnect): new function
825 2003-09-21 Havoc Pennington <hp@pobox.com>
827 Get matching rules mostly working in the bus; only actually
828 parsing the rule text remains. However, the client side of
829 "signal connections" hasn't been started, this patch is only the
832 * dbus/dispatch.c: fix for the matching rules changes
834 * bus/driver.c (bus_driver_handle_remove_match)
835 (bus_driver_handle_add_match): send an ack reply from these
838 * glib/dbus-gproxy.c (dbus_gproxy_begin_call): fix order of
839 arguments, reported by Seth Nickell
841 * bus/config-parser.c (append_rule_from_element): support
842 eavesdrop=true|false attribute on policies so match rules
843 can be prevented from snooping on the system bus.
845 * bus/dbus-daemon-1.1.in: consistently use terminology "sender"
846 and "destination" in attribute names; fix some docs bugs;
847 add eavesdrop=true|false attribute
849 * bus/driver.c (bus_driver_handle_add_match)
850 (bus_driver_handle_remove_match): handle AddMatch, RemoveMatch
853 * dbus/dbus-protocol.h (DBUS_SERVICE_ORG_FREEDESKTOP_BROADCAST): get
854 rid of broadcast service concept, signals are just always broadcast
856 * bus/signals.c, bus/dispatch.c, bus/connection.c, bus/bus.c:
857 mostly implement matching rules stuff (currently only exposed as signal
860 2003-09-21 Mark McLoughlin <mark@skynet.ie>
862 * doc/dbus-specification.sgml: Change the header field name
863 to be an enum and update the rest of the spec to reference
864 the fields using the conventinal name.
866 * dbus/dbus-protocol.h: update to reflect the spec.
868 * doc/TODO: add item to remove the 4 byte alignment requirement.
870 * dbus/dbus-message.c: Remove the code to generalise the
871 header/body length and serial number header fields as named
872 header fields so we can reference field names using the
874 (append_int_field), (append_uint_field), (append_string_field):
875 Append the field name as a byte rather than four chars.
876 (delete_int_or_uint_field), (delete_string_field): reflect the
877 fact that the field name and typecode now occupy 4 bytes instead
879 (decode_string_field), (decode_header_data): update to reflect
880 protocol changes and move the field specific encoding from
881 decode_string_field() back into decode_header_data().
883 * dbus/dbus-internals.[ch]: (_dbus_header_field_to_string):
884 Add utility to aid debugging.
886 * dbus/dbus-message-builder.c:
887 (append_string_field), (_dbus_message_data_load): Update to
888 reflect protocol changes; Change the FIELD_NAME directive
889 to HEADER_FIELD and allow it to take the field's conventional
890 name rather than the actual value.
892 * test/data/*/*.message: Update to use HEADER_FIELD instead
893 of FIELD_NAME; Always align the header on an 8 byte boundary
894 *before* updating the header length.
896 2003-09-15 Havoc Pennington <hp@pobox.com>
898 * dbus/dbus-pending-call.c: add the get/set object data
899 boilerplate as for DBusConnection, etc. Use generic object data
900 for the notify callback.
902 * glib/dbus-gparser.c (parse_node): parse child nodes
904 * tools/dbus-viewer.c: more hacking on the dbus-viewer
906 * glib/dbus-gutils.c (_dbus_gutils_split_path): add a file to
907 contain functions shared between the convenience lib and the
910 * glib/Makefile.am (libdbus_glib_1_la_LDFLAGS): add
911 -export-symbols-regex to the GLib library
913 * dbus/dbus-object-tree.c (_dbus_object_tree_dispatch_and_unlock):
914 fix the locking in here, and add a default handler for
915 Introspect() that just returns sub-nodes.
917 2003-09-14 Havoc Pennington <hp@pobox.com>
919 * glib/dbus-gthread.c (dbus_g_thread_init): rename to make g_foo
920 rather than gfoo consistent
922 * glib/dbus-gproxy.h: delete for now, move contents to
923 dbus-glib.h, because the include files don't work right since we
924 aren't in the dbus/ subdir.
926 * glib/dbus-gproxy.c (dbus_gproxy_send): finish implementing
927 (dbus_gproxy_end_call): finish
928 (dbus_gproxy_begin_call): finish
930 * glib/dbus-gmain.c (dbus_set_g_error): new
932 * glib/dbus-gobject.c (handle_introspect): include information
933 about child nodes in the introspection
935 * dbus/dbus-connection.c (dbus_connection_list_registered): new
936 function to help in implementation of introspection
938 * dbus/dbus-object-tree.c
939 (_dbus_object_tree_list_registered_and_unlock): new function
941 2003-09-12 Havoc Pennington <hp@pobox.com>
943 * glib/dbus-gidl.h: add common base class for all the foo_info
946 * tools/dbus-viewer.c: add GTK-based introspection UI thingy
949 * test/Makefile.am: try test srcdir -ef . in addition to test
950 srcdir = ., one of them should work (yeah lame)
952 * glib/Makefile.am: build the "idl" parser stuff as a convenience
955 * glib/dbus-gparser.h: make description_load routines return
956 NodeInfo* not Parser*
958 * Makefile.am (SUBDIRS): build test dir after all library dirs
960 * configure.in: add GTK+ detection
962 2003-09-07 Havoc Pennington <hp@pobox.com>
964 * Make Doxygen contented.
966 2003-09-07 Havoc Pennington <hp@pobox.com>
968 * doc/dbus-specification.sgml: more updates
970 2003-09-06 Havoc Pennington <hp@pobox.com>
972 * doc/dbus-specification.sgml: partial updates
974 * bus/dbus-daemon-1.1.in: fix the config file docs for the
975 zillionth time; hopefully I edited the right file this time.
977 * bus/config-parser.c (append_rule_from_element): support
978 send_type, send_path, receive_type, receive_path
980 * bus/policy.c: add message type and path to the list of things
981 that can be "firewalled"
983 2003-09-06 Havoc Pennington <hp@pobox.com>
985 * dbus/dbus-connection.c (dbus_connection_register_fallback): add this
986 (dbus_connection_register_object_path): make this not handle
987 messages to paths below the given path
989 2003-09-03 Havoc Pennington <hp@pobox.com>
991 * test/glib/Makefile.am: add this with random glib-linked test
994 * glib/Makefile.am: remove the random test programs from here,
995 leave only the unit tests
997 * glib/dbus-gobject.c (_dbus_gobject_test): add test for
998 uscore/javacaps conversion, and fix
999 (get_object_property, set_object_property): change to .NET
1000 convention for mapping props to methods, set_FooBar/get_FooBar,
1001 since one language has such a convention we may as well copy it.
1002 Plus real methods in either getFooBar or get_foo_bar style won't
1003 collide with this convention.
1005 2003-09-01 Havoc Pennington <hp@pobox.com>
1007 * glib/dbus-gparser.c: implement
1009 * glib/dbus-gobject.c: start implementing skeletons support
1011 * configure.in: when disabling checks/assert, also define
1012 G_DISABLE_ASSERT and G_DISABLE_CHECKS
1014 2003-09-01 Havoc Pennington <hp@pobox.com>
1016 * glib/Makefile.am: rearrange a bunch of files and get "make
1017 check" framework set up
1019 2003-08-31 Havoc Pennington <hp@pobox.com>
1021 * fix build with --disable-tests
1023 2003-08-30 Havoc Pennington <hp@pobox.com>
1025 * dbus/dbus-connection.c: purge DBusMessageHandler
1027 * dbus/dbus-message-handler.c: remove DBusMessageHandler, just
1028 use callbacks everywhere
1030 2003-08-30 Havoc Pennington <hp@pobox.com>
1032 * test/data/valid-config-files/system.d/test.conf: change to
1033 root for the user so warnings don't get printed
1035 * dbus/dbus-message.c: add dbus_message_get_path,
1036 dbus_message_set_path
1038 * dbus/dbus-object-tree.c (do_test_dispatch): add test of
1039 dispatching to a path
1041 * dbus/dbus-string.c (_dbus_string_validate_path): add
1043 * dbus/dbus-marshal.c (_dbus_demarshal_object_path): implement
1044 (_dbus_marshal_object_path): implement
1046 * dbus/dbus-protocol.h (DBUS_HEADER_FIELD_PATH): new header field
1047 to contain the path to the target object
1048 (DBUS_HEADER_FIELD_SENDER_SERVICE): rename
1049 DBUS_HEADER_FIELD_SENDER to explicitly say it's the sender service
1051 2003-08-30 Havoc Pennington <hp@pobox.com>
1053 * dbus/dbus-object-tree.c: write tests and fix the discovered bugs
1055 2003-08-29 Havoc Pennington <hp@pobox.com>
1057 * dbus/dbus-object-tree.c: modify to allow overlapping paths to be
1059 (struct DBusObjectSubtree): shrink this
1060 a lot, since we may have a lot of them
1061 (_dbus_object_tree_free_all_unlocked): implement
1062 (_dbus_object_tree_dispatch_and_unlock): implement
1064 2003-08-29 Havoc Pennington <hp@pobox.com>
1066 * dbus/dbus-internals.h: fix _DBUS_N_GLOBAL_LOCKS
1068 2003-08-28 Havoc Pennington <hp@pobox.com>
1072 * dbus/dbus-connection.c: port to no ObjectID, create a
1073 DBusObjectTree, rename ObjectTree to ObjectPath in public API
1075 * dbus/dbus-connection.h (struct DBusObjectTreeVTable): delete
1076 everything except UnregisterFunction and MessageFunction
1078 * dbus/dbus-marshal.c: port away from DBusObjectID,
1079 add DBUS_TYPE_OBJECT_PATH
1081 * dbus/dbus-object-registry.[hc], dbus/dbus-object.[hc],
1082 dbus/dbus-objectid.[hc]: remove these, we are moving to
1083 path-based object IDs
1085 2003-08-25 Havoc Pennington <hp@pobox.com>
1087 Just noticed that dbus_message_test is hosed, I wonder when I
1088 broke that. I thought make check was passing earlier...
1090 * dbus/dbus-object-tree.c: add new "object tree" to match DCOP
1091 container tree, will replace most of dbus-object-registry
1093 * dbus/dbus-string.c (_dbus_string_append_printf_valist): fix C99
1096 2003-08-19 Havoc Pennington <hp@pobox.com>
1098 * dbus/dbus-message.c (decode_string_field): support FIELD_SENDER
1099 (dbus_message_is_error): fix this function
1101 * bus/dbus-daemon-1.1: clarify logic on when <deny>/<allow> rules
1104 * bus/policy.c (bus_client_policy_check_can_receive): fix code to
1105 reflect clarified man page
1106 (bus_client_policy_check_can_send): ditto
1108 * bus/session.conf.in: fixup
1110 * bus/system.conf.in: fixup
1112 2003-08-18 Havoc Pennington <hp@redhat.com>
1114 * dbus/dbus-hash.c (_dbus_hash_table_insert_two_strings): fix
1116 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
1117 dumb bug created earlier (wrong order of args to
1118 decode_header_data())
1120 * tools/dbus-send.c: port
1122 * tools/dbus-print-message.c (print_message): port
1124 * test/data/*messages: port all messages over
1126 * dbus/dbus-message-builder.c: support including
1129 * bus/driver.c: port over
1131 * bus/dispatch.c: port over to new stuff
1133 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
1134 rename disconnect signal to "Disconnected"
1136 2003-08-17 Havoc Pennington <hp@pobox.com>
1138 This doesn't compile yet, but syncing up so I can hack on it from
1139 work. What are branches for if not broken code? ;-)
1141 * dbus/dbus-protocol.h: remove DBUS_HEADER_FIELD_NAME, add
1142 DBUS_HEADER_FIELD_INTERFACE, DBUS_HEADER_FIELD_MEMBER,
1143 DBUS_HEADER_FIELD_ERROR_NAME
1145 * dbus/dbus-hash.c: Introduce DBUS_HASH_TWO_STRINGS as hack to use
1146 for the interface+member pairs
1147 (string_hash): change to use g_str_hash algorithm
1148 (find_direct_function, find_string_function): refactor these to
1151 * dbus/dbus-message.c: port all of this over to support
1152 interface/member fields instead of name field
1154 * dbus/dbus-object-registry.c: port over
1156 * dbus/dbus-string.c (_dbus_string_validate_interface): rename
1157 from _dbus_string_validate_name
1159 * bus/dbus-daemon-1.1: change file format for the
1160 <deny>/<allow> stuff to match new message naming scheme
1162 * bus/policy.c: port over
1164 * bus/config-parser.c: parse new format
1166 2003-08-16 Havoc Pennington <hp@pobox.com>
1168 * dbus/dbus-object-registry.c (add_and_remove_objects): remove
1171 * glib/dbus-gproxy.c: some hacking
1173 2003-08-15 Havoc Pennington <hp@redhat.com>
1175 * dbus/dbus-pending-call.c (dbus_pending_call_block): implement
1177 * dbus/dbus-connection.c
1178 (dbus_connection_send_with_reply_and_block): factor out internals;
1179 change to convert any error replies to DBusError instead of
1180 returning them as a message
1182 2003-08-15 Havoc Pennington <hp@pobox.com>
1184 * dbus/dbus-connection.c,
1185 dbus/dbus-pending-call.c: Finish the pending call stuff
1187 2003-08-14 Havoc Pennington <hp@redhat.com>
1189 * dbus/dbus-pending-call.c: start on new object that will replace
1190 DBusMessageHandler and ReplyHandlerData for tracking outstanding
1193 * dbus/dbus-gproxy.c: start on proxy object used to communicate
1194 with remote interfaces
1196 * dbus/dbus-gidl.c: do the boring boilerplate in here
1198 2003-08-12 Havoc Pennington <hp@pobox.com>
1200 * bus/dispatch.c (bus_dispatch): make this return proper
1201 DBusHandlerResult to avoid DBUS_ERROR_UNKNOWN_METHOD
1203 * dbus/dbus-errors.c (dbus_set_error): use
1204 _dbus_string_append_printf_valist
1206 * dbus/dbus-string.c (_dbus_string_append_printf_valist)
1207 (_dbus_string_append_printf): new
1209 * dbus/dbus-errors.h (DBUS_ERROR_UNKNOWN_MESSAGE): change to
1212 * dbus/dbus-connection.c (dbus_connection_dispatch): handle
1213 DBUS_HANDLER_RESULT_NEED_MEMORY; send default error reply if a
1214 message is unhandled.
1216 2003-08-11 Havoc Pennington <hp@pobox.com>
1218 * bus/test.c (client_disconnect_handler): change to return
1219 HANDLED (would have been REMOVE_MESSAGE)
1221 * dbus/dbus-object.h (enum DBusHandlerResult): rename to
1222 HANDLED/NOT_YET_HANDLED instead of
1223 REMOVE_MESSAGE/ALLOW_MORE_HANDLERS to make it clearer how it
1226 2003-08-10 Havoc Pennington <hp@pobox.com>
1228 * tools/dbus-send.c (main): add --type argument, for now
1229 supporting only method_call and signal types.
1231 * tools/dbus-print-message.c: print message type
1233 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
1234 init connection->objects
1236 * doc/dbus-specification.sgml: fix sgml
1238 * bus/*.c: port over to object-instance API changes
1240 * test/test-service.c: ditto
1242 * dbus/dbus-message.c (dbus_message_create_header): allow #NULL
1243 name, we will have to fix up the rest of the code to also handle
1245 (dbus_message_new): generic message-creation call
1246 (set_string_field): allow appending name field
1248 2003-08-06 Havoc Pennington <hp@pobox.com>
1250 * dbus/dbus-object-registry.c: implement signal connection
1253 * dbus/dbus-connection.c (_dbus_connection_unref_unlocked): new
1255 * dbus/dbus-internals.c (_dbus_memdup): new function
1257 2003-08-02 Havoc Pennington <hp@pobox.com>
1259 * dbus/dbus-message.c (dbus_message_get_no_reply)
1260 (dbus_message_set_no_reply): add these and remove
1261 set_is_error/get_is_error
1263 * dbus/dbus-protocol.h, doc/dbus-specification.sgml:
1264 remove the ERROR flag, since there's now an ERROR type
1266 2003-08-01 Havoc Pennington <hp@pobox.com>
1268 * dbus/dbus-object-registry.c (_dbus_object_registry_handle_and_unlock):
1271 * dbus/dbus-message.c (dbus_message_get_type): new function
1273 * doc/dbus-specification.sgml: add "type" byte to messages
1275 2003-08-01 Havoc Pennington <hp@pobox.com>
1277 * dbus/dbus-protocol.h (DBUS_MESSAGE_TYPE_*): introduce
1278 a message type enum to distinguish kinds of message
1279 (DBUS_HEADER_FLAG_NO_REPLY_EXPECTED): flag for a message
1280 that need not be replied to
1282 2003-08-01 Havoc Pennington <hp@pobox.com>
1284 * dbus/dbus-marshal.c: adapt to DBusObjectID changes
1285 (unpack_8_octets): fix no-64-bit-int bug
1287 * dbus/dbus-object-registry.c (validate_id): validate the
1288 connection ID bits, not just the instance ID.
1290 * dbus/dbus-connection.c (_dbus_connection_init_id): initialize
1291 the connection-global 33 bits of the object ID
1293 * dbus/dbus-object-registry.c (info_from_entry): fill in
1294 object ID in the new way
1296 * dbus/dbus-objectid.h: rather than high/low bits, specifically
1297 define server/client/instance bits.
1299 2003-07-30 Havoc Pennington <hp@pobox.com>
1301 * dbus/dbus-connection.c (dbus_connection_register_object): fix
1304 2003-07-13 Havoc Pennington <hp@pobox.com>
1306 * dbus/dbus-object.h (struct DBusObjectVTable): add padding
1307 fields to DBusObjectVTable and DBusObjectInfo
1309 2003-07-12 Havoc Pennington <hp@pobox.com>
1311 * dbus/dbus-object-registry.c: implement unit test,
1312 fix bugs discovered in process
1314 * dbus/dbus-connection.c: remove handler_table and
1315 register_handler(), add DBusObjectRegistry usage
1317 * dbus/dbus-objectid.c (dbus_object_id_is_null)
1318 (dbus_object_id_set_null): new functions
1320 2003-07-08 Havoc Pennington <hp@pobox.com>
1322 * dbus/dbus-object.c: implement some of this
1324 * dbus/dbus-object-registry.c
1325 (_dbus_object_registry_add_and_unlock): fill in the object_id out
1327 (_dbus_object_registry_new): handle OOM
1329 2003-07-08 Havoc Pennington <hp@pobox.com>
1331 * dbus/dbus-object.h: sketch out an API for registering objects
1332 with a connection, that allows us to use as little as 24 bytes
1333 per object and lets application code represent an object in
1334 any conceivable way.
1336 * dbus/dbus-object-registry.c: implement the hard bits of the
1337 DBusConnection aspect of object API. Not yet wired up.
1339 2003-07-06 Havoc Pennington <hp@pobox.com>
1341 * dbus/dbus-marshal.c (_dbus_marshal_set_object_id): new function
1342 (_dbus_marshal_object_id): new
1343 (_dbus_demarshal_object_id): new
1344 (_dbus_marshal_get_arg_end_pos): support object ID type, and
1345 consolidate identical switch cases. Don't conditionalize handling
1346 of DBUS_TYPE_UINT64, need to handle the type always.
1347 (_dbus_marshal_validate_arg): consolidate identical cases, and
1348 handle DBUS_TYPE_OBJECT_ID
1350 * dbus/dbus-objectid.c: new file with DBusObjectID data type.
1352 * dbus/dbus-protocol.h: add DBUS_TYPE_OBJECT_ID
1354 2003-09-28 Havoc Pennington <hp@pobox.com>
1358 2003-09-28 Havoc Pennington <hp@pobox.com>
1360 * doc/Makefile.am (dbus-specification.html): testing a funky hack
1361 to work with Debian db2html
1363 2003-09-28 Havoc Pennington <hp@pobox.com>
1365 * configure.in: 0.13
1367 * doc/Makefile.am (dbus-test-plan.html): accept nonexistence of
1368 stylesheet-images for benefit of Debian
1370 Change back to using filesystem-linked sockets for the system
1371 bus, so only root can create the default system bus address.
1373 * bus/system.conf.in: change to use
1374 DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
1376 * dbus/Makefile.am (INCLUDES): remove DBUS_SYSTEM_BUS_PATH define
1379 * configure.in: define DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
1380 here, and AC_DEFINE DBUS_SYSTEM_PATH
1382 2003-08-09 Anders Carlsson <andersca@codefactory.se>
1385 * doc/busconfig.dtd:
1388 2003-08-09 Anders Carlsson <andersca@codefactory.se>
1390 * doc/dbus-specification.sgml:
1391 Add activation reply values.
1393 2003-08-05 Havoc Pennington <hp@redhat.com>
1395 * configure.in: 0.12
1397 2003-08-05 Anders Carlsson <andersca@codefactory.se>
1399 * glib/dbus-gmain.c: (watch_fd_new), (watch_fd_ref),
1400 (watch_fd_unref), (dbus_gsource_check), (dbus_gsource_dispatch),
1401 (add_watch), (remove_watch), (create_source):
1402 Refcount fds, fixes some reentrancy issues.
1404 2003-07-30 Havoc Pennington <hp@redhat.com>
1406 * dbus/dbus-bus.c (init_connections_unlocked): fix default system
1407 bus address to be abstract if we have abstract sockets
1411 2003-07-28 Havoc Pennington <hp@redhat.com>
1413 * bus/messagebus.in: fix to avoid processname/servicename
1414 confusion, from Michael Kearey
1415 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=100965
1417 2003-07-23 Havoc Pennington <hp@pobox.com>
1419 * dbus/dbus-message.c (dbus_message_iter_get_named):
1420 fix from Andy Hanton to remove broken "+1"
1422 2003-07-16 Havoc Pennington <hp@pobox.com>
1424 * tools/dbus-launch.c (babysit): close stdout/stderr in the
1425 babysitter process, as suggested by Thomas Leonard, so
1426 an "eval `dbus-launch --exit-with-session`" will actually
1429 2003-07-16 Havoc Pennington <hp@pobox.com>
1431 * configure.in: print out EXPANDED_* variables in the summary at
1432 the end; clean up the code that computes EXPANDED_ variables and
1433 get the ones using exec_prefix right. Should make things work
1434 when you build without --prefix
1436 2003-06-29 Havoc Pennington <hp@pobox.com>
1438 * mono/Test.cs (class Test): fire up a main loop and run it
1440 * mono/DBus.cs (DBus): don't g_thread_init since it can only be
1441 done once, the app has to do it
1443 2003-06-26 Havoc Pennington <hp@pobox.com>
1445 * mono/Connection.cs: set up connection with the glib main loop
1447 2003-07-01 Havoc Pennington <hp@redhat.com>
1449 * doc/dbus-specification.sgml: clarify the format of a type code,
1450 change suggested by Jim Blandy
1452 2003-06-29 Miloslav Trmac <mitr@volny.cz>
1455 * tools/Makefile.am: Don't assume srcdir == builddir.
1457 * dbus/dbus-memory.c (dbus_realloc): Don't check guards after shrinking
1458 the allocated block.
1459 (_dbus_memory_test): New function.
1460 * dbus/dbus-test.h: Add _dbus_memory_test ().
1461 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): Call it.
1463 * dbus/dbus-message.c (decode_header_data): Use %.4s instead
1465 (dbus_message_new): Remove obsolete @todo.
1467 * dbus/dbus-marshal.c (_dbus_marshal_set_int64)
1468 (_dbus_marshal_set_uint64): Fix comment.
1470 * dbus/dbus-message.c (append_int_field, append_uint_field): Don't
1471 hardcode FIELD_REPLY_SERIAL.
1473 * dbus/dbus-mainloop.c (_dbus_loop_remove_watch)
1474 (_dbus_loop_remove_timeout): Cast function pointers to (void *) for %p
1476 * configure.in: Add -D_POSIX_C_SOURCE=199309L -DBSD_SOURCE to CFLAGS
1477 and disable DBUS_USE_ATOMIC_INT_486 when --enable-ansi is used
1479 2003-06-24 Havoc Pennington <hp@pobox.com>
1481 * mono/*.cs: Use IntPtr.Zero instead of ((IntPtr) 0)
1483 2003-06-23 Anders Carlsson <andersca@codefactory.se>
1489 * gcj/TestMessage.java: (TestMessage), (TestMessage.main):
1490 * gcj/org/.cvsignore:
1491 * gcj/org/Makefile.am:
1492 * gcj/org/freedesktop/.cvsignore:
1493 * gcj/org/freedesktop/Makefile.am:
1494 * gcj/org/freedesktop/dbus/.cvsignore:
1495 * gcj/org/freedesktop/dbus/Makefile.am:
1496 * gcj/org/freedesktop/dbus/Message.java: (Message),
1498 * gcj/org/freedesktop/dbus/natMessage.cc:
1499 Fix the build system.
1501 2003-06-22 Havoc Pennington <hp@pobox.com>
1503 * mono/Connection.cs: add more bindings
1505 * dbus/dbus-threads.c (dbus_threads_init): allow calling this
1508 2003-06-22 Havoc Pennington <hp@pobox.com>
1510 * mono/Connection.cs, mono/DBus.cs, mono/Error.cs:
1511 Start wrapping more stuff.
1513 2003-06-22 Havoc Pennington <hp@pobox.com>
1515 * mono/Message.cs: implement Message.Wrap() that ensures we only
1516 have a single C# wrapper per DBusMessage, assuming it works which
1517 it probably doesn't.
1519 * dbus/dbus-message.c (dbus_message_allocate_data_slot): new
1520 (dbus_message_free_data_slot): new
1521 (dbus_message_set_data): new
1522 (dbus_message_get_data): new
1524 2003-06-22 Havoc Pennington <hp@pobox.com>
1526 * dbus/dbus-dataslot.c (_dbus_data_slot_allocator_unref)
1527 (_dbus_data_slot_allocator_alloc): rework these to keep a
1528 reference count on each slot and automatically manage a global
1529 slot ID variable passed in by address
1531 * bus/bus.c: convert to new dataslot API
1533 * dbus/dbus-bus.c: convert to new dataslot API
1535 * dbus/dbus-connection.c: convert to new dataslot API
1537 * dbus/dbus-server.c: convert to new dataslot API
1539 * glib/dbus-gmain.c: ditto
1543 * bus/connection.c: ditto
1545 2003-06-22 Anders Carlsson <andersca@codefactory.se>
1547 * configure.in: Add AM_PROG_GCJ and move AM_PROG_LIBTOOL
1548 after the gcj checks so that the correct configuration tags
1549 will be added to libtool.
1551 * dbus-glib-1.pc.in: No need to specify any includes since
1552 dbus-1.pc.in has those.
1554 2003-06-22 Havoc Pennington <hp@pobox.com>
1556 * mono/*, gcj/*, configure.in, Makefile.am:
1557 Check in makefiles and subdirs for mono and gcj bindings.
1558 Neither binding actually exists, just trying to get through
1559 all the build and other boring bits.
1561 2003-06-21 Philip Blundell <philb@gnu.org>
1563 * tools/dbus-monitor.1: Updated.
1565 * tools/dbus-send.1: Likewise.
1567 2003-06-20 Anders Carlsson <andersca@codefactory.se>
1569 * dbus/dbus-transport-unix.c (unix_handle_watch): Check
1570 for hangup and error after checking read so we won't discard
1571 pending data if both hangup and read are set.
1573 2003-06-19 Philip Blundell <philb@gnu.org>
1575 * tools/dbus-print-message.c (print_message): Handle BOOLEAN.
1577 * tools/dbus-send.c: Accept both --system and --session.
1579 * tools/dbus-monitor.c: Same here.
1581 2003-06-19 Anders Carlsson <andersca@codefactory.se>
1583 * glib/dbus-glib.h: Fix so that dbus-glib.h can be used
1584 from C++ (Patch by Miloslav Trmac).
1586 2003-06-15 Joe Shaw <joe@assbarn.com>
1588 * configure.in: Check for socklen_t.
1590 * dbus/dbus-sysdeps.c: Define socklen_t if it's not defined.
1592 * test/test-segfault.c: Add #include <sys/time.h>
1594 * tools/Makefile.am: Add DBUS_X_CFLAGS to the INCLUDES since
1595 dbus-launch needs it.
1597 2003-06-09 Havoc Pennington <hp@redhat.com>
1599 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): don't use
1600 SUN_LEN, it breaks abstract socket usage
1602 * dbus/dbus-internals.c (_dbus_verbose_real): only print PID at
1605 2003-06-04 Havoc Pennington <hp@pobox.com>
1607 * dbus/dbus-server.c (dbus_server_listen): allow abstract sockets
1608 using unix:abstract=/foo, and when listening in a tmpdir
1609 i.e. unix:tmpdir=/tmp, always use abstract sockets if we can.
1611 * dbus/dbus-transport.c (_dbus_transport_open): support
1614 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
1615 support abstract sockets
1617 * dbus/dbus-transport-unix.c
1618 (_dbus_transport_new_for_domain_socket): support abstract sockets
1620 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket): add "abstract"
1621 toggle as an argument, implement abstract namespace support
1622 (_dbus_listen_unix_socket): ditto
1624 * configure.in: add --enable-abstract-sockets and implement
1625 a configure check for autodetection of the right value.
1627 2003-06-01 Havoc Pennington <hp@pobox.com>
1629 * tools/dbus-cleanup-sockets.c: add utility to clean up sockets
1630 in /tmp (though on Linux this will end up being useless,
1631 when we add abstract namespace support)
1633 * configure.in: define DBUS_SESSION_SOCKET_DIR in addition to
1636 2003-05-28 Colin Walters <walters@verbum.org>
1638 * tools/dbus-monitor.c (main): Fix silly typo (s/--session/--system/).
1640 2003-05-18 Anders Carlsson <andersca@codefactory.se>
1642 * dbus/dbus-message.c (dbus_message_new): Remove @todo.
1644 2003-05-17 Colin Walters <walters@gnu.org>
1646 * tools/dbus-send.c: Don't exit with an error code if --help was
1647 passed. Default to using the session bus instead of the system
1650 * tools/dbus-launch.c: Ditto.
1652 * tools/dbus-monitor.c: Ditto.
1654 * tools/dbus-send.1: Update with new arguments.
1656 * tools/dbus-launch.c: Emit code to export variables. New
1657 arguments -s and -c to specify shell syntax, and a bit of code to
1658 autodetect syntax. Also, allow specifying a program to run.
1660 * tools/dbus-launch.1: Update with new arguments.
1662 * tools/dbus-send.1: Ditto.
1664 * tools/dbus-monitor.1: Ditto.
1666 2003-05-17 Havoc Pennington <hp@pobox.com>
1668 * bus/config-parser.c (merge_included): merge in policies from
1669 child configuration file.
1671 * bus/policy.c (bus_policy_merge): function to merge two policies
1674 2003-05-16 Havoc Pennington <hp@redhat.com>
1676 * dbus/dbus-connection.c: disable verbose lock spew
1678 * tools/dbus-send.c: add --print-reply command line option
1680 * tools/dbus-print-message.h (print_message): new util function
1681 shared by dbus-send and dbus-monitor
1683 * tools/dbus-monitor.c (handler_func): exit on disconnect
1685 * dbus/dbus-transport-unix.c (do_reading): if the transport is
1686 disconnected, don't try to use the read_watch
1688 * dbus/dbus-watch.c (dbus_watch_get_enabled): assert watch != NULL
1689 so we can find this bug more easily
1691 2003-05-16 Havoc Pennington <hp@redhat.com>
1693 * bus/policy.c (free_rule_list_func): avoid a crash when passed
1694 NULL as DBusHashTable is annoyingly likely to do.
1696 2003-05-16 Colin Walters <walters@verbum.org>
1698 * tools/dbus-monitor.c: Add --session argument and usage()
1701 * tools/dbus-monitor.1: Update with new --session arg.
1703 * bus/Makefile.am (install-data-hook): Create
1704 $(libdir)/dbus-1.0/services so that the session bus is happy.
1706 2003-05-15 Havoc Pennington <hp@redhat.com>
1708 * dbus/dbus-sysdeps.c (_dbus_atomic_dec, _dbus_atomic_inc): work
1709 on non-x86. ifdef's are evil.
1711 2003-05-15 Havoc Pennington <hp@redhat.com>
1713 * configure.in: 0.11
1717 * bus/Makefile.am (initddir): apparently we are supposed to put
1718 init scripts in /etc/rc.d/init.d not /etc/init.d
1720 * bus/Makefile.am: remove the "you must --enable-tests to make
1721 check" as it broke distcheck
1723 * bus/Makefile.am (install-data-hook): create /etc/dbus-1/system.d
1725 2003-05-13 James Willcox <jwillcox@gnome.org>
1728 * bus/activation.c: (bus_activation_service_created),
1729 (bus_activation_activate_service):
1730 * bus/driver.c: (bus_driver_send_service_deleted),
1731 (bus_driver_send_service_created), (bus_driver_send_service_lost),
1732 (bus_driver_send_service_acquired),
1733 (bus_driver_send_welcome_message),
1734 (bus_driver_handle_list_services):
1735 * bus/session.conf.in:
1736 * dbus/dbus-bus.c: (dbus_bus_acquire_service),
1737 (dbus_bus_service_exists), (dbus_bus_activate_service):
1740 Add some convenience API which lets you activate a service, and did a
1741 bunch of s/0/DBUS_TYPE_INVALID/ in calls to dbus_message_append_args()
1742 and dbus_message_get_args()
1744 2003-05-11 Havoc Pennington <hp@pobox.com>
1746 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix to avoid
1747 calling _dbus_marshal_validate_arg() for every byte in a byte
1750 * dbus/dbus-message-handler.c: use atomic reference counting to
1751 reduce number of locks slightly; the global lock in here sucks
1753 * dbus/dbus-connection.c
1754 (_dbus_connection_update_dispatch_status_and_unlock): variant of
1755 update_dispatch_status that can be called with lock held; then use
1756 in a couple places to reduce locking/unlocking
1757 (dbus_connection_send): hold the lock over the whole function
1758 instead of acquiring it twice.
1760 * dbus/dbus-timeout.c (_dbus_timeout_new): handle OOM
1762 * bus/connection.c (bus_connections_setup_connection): fix access
1763 to already-freed memory.
1765 * dbus/dbus-connection.c: keep a little cache of linked list
1766 nodes, to avoid using the global linked list alloc lock in the
1767 normal send-message case. Instead we just use the connection lock
1768 that we already have to take.
1770 * dbus/dbus-list.c (_dbus_list_find_last): new function
1772 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec):
1773 change to use a struct for the atomic type; fix docs,
1774 they return value before increment, not after increment.
1776 * dbus/dbus-string.c (_dbus_string_append_4_aligned)
1777 (_dbus_string_append_8_aligned): new functions to try to
1778 microoptimize this operation.
1779 (reallocate_for_length): break this out of set_length(), to
1780 improve profile info, and also so we can consider inlining the
1783 * dbus/dbus-message.c (dbus_message_new_empty_header): init data
1784 strings with some preallocation, cuts down on our calls to realloc
1785 a fair bit. Though if we can get the "move entire string to empty
1786 string" optimization below to kick in here, it would be better.
1788 * dbus/dbus-string.c (_dbus_string_move): just call
1789 _dbus_string_move_len
1790 (_dbus_string_move_len): add a special case for moving
1791 an entire string into an empty string; we can just
1792 swap the string data instead of doing any reallocs.
1793 (_dbus_string_init_preallocated): new function
1795 2003-05-11 Havoc Pennington <hp@pobox.com>
1797 Write a "test-profile" that does echo client-server with threads;
1798 profile reveals lock contention, memcpy/realloc of buffers, and
1799 UTF-8 validation as hot spots. 20% of lock contention eliminated
1800 with dbus_atomic_inc/dec implementation on x86. Much remaining
1801 contention is global mempool locks for GList and DBusList.
1803 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec): add
1806 * dbus/dbus-connection.c (struct DBusConnection): use
1807 dbus_atomic_t for the reference count
1809 * dbus/dbus-message.c (struct DBusMessage): declare
1810 dbus_atomic_t values as volatile
1812 * configure.in: code to detect ability to use atomic integer
1813 operations in assembly, from GLib patch
1815 * dbus/dbus-internals.c (_dbus_verbose_real): call getpid every
1816 time, tired of it being wrong in threads and forked processes
1818 * glib/test-profile.c: a little program to bounce messages back
1819 and forth between threads and eat CPU
1821 * dbus/dbus-connection.c: add debug spew macros for debugging
1822 thread locks; include config.h at top; fix deadlock in
1823 dbus_connection_flush()
1825 2003-05-08 Havoc Pennington <hp@pobox.com>
1827 * dbus/dbus-spawn.c: s/_exit/exit/ because it was keeping gcov
1828 data from getting written, and there wasn't a good reason to
1831 * test/decode-gcov.c (mark_inside_dbus_build_tests): don't count
1832 dbus_verbose lines in test coverage
1833 (main): add list of functions sorted by # of untested blocks
1834 to the coverage report
1836 * dbus/dbus-mempool.c: put some test-only code in DBUS_BUILD_TESTS
1838 * dbus/dbus-marshal.c (_dbus_marshal_test): extend test coverage
1840 * dbus/dbus-message-handler.c (_dbus_message_handler_test):
1841 extend test coverage
1843 * test/data/auth/cancel.auth-script: test canceling an
1846 * dbus/Makefile.am: remove dbus-server-debug.[hc] for now, as they
1847 aren't used. in CVS history if we end up needing them.
1849 2003-05-04 Havoc Pennington <hp@pobox.com>
1851 * dbus/dbus-message-handler.c (_dbus_message_handler_test): add
1854 * dbus/dbus-marshal.c (_dbus_demarshal_string_array): fix this
1855 function, which assumed length was in # of strings, not bytes
1857 * dbus/dbus-message.c (_dbus_message_test): add tests for some
1860 * dbus/dbus-connection.c
1861 (_dbus_connection_queue_received_message): disable function for
1862 now, we are only using it in test mode
1864 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
1865 remove a mistaken FIXME
1867 2003-05-04 Havoc Pennington <hp@pobox.com>
1869 * dbus/dbus-connection.c (dbus_connection_preallocate_send):
1870 unlock mutex on successful return, patch from Anders Gustafsson
1872 2003-05-04 Havoc Pennington <hp@pobox.com>
1874 * dbus-glib-1.pc.in (Requires): fix dependencies, from
1877 2003-05-04 Havoc Pennington <hp@pobox.com>
1879 * tools/dbus-launch.c: implement
1881 * bus/main.c (main), bus/bus.c (bus_context_new):
1882 implement --print-pid and --fork
1884 2003-05-03 Havoc Pennington <hp@redhat.com>
1886 * dbus/dbus-address.c (dbus_parse_address): fix bug when a key in
1887 the address had no value, and add to test suite. Fix and
1888 regression test from Miloslav Trmac
1890 2003-05-03 Havoc Pennington <hp@pobox.com>
1892 * dbus/dbus-watch.c (dbus_watch_handle): warn and return if a
1893 watch is invalid when handled
1895 * tools/Makefile.am, tools/dbus-launch.c, tools/dbus-launch.1: add
1896 dbus-launch utility to launch the bus from a shell script. Didn't
1897 actually implement dbus-launch yet, it's just a placeholder still.
1899 2003-05-03 Havoc Pennington <hp@pobox.com>
1901 * bus/Makefile.am, bus/dbus-daemon-1.1.in: man page for the
1902 daemon; also documents daemon config file, so replaces
1903 doc/config-file.txt. Corrected some stuff from config-file.txt in
1904 the process of moving it.
1906 2003-05-03 Havoc Pennington <hp@pobox.com>
1908 * tools/Makefile.am, tools/dbus-send.1, tools/dbus-monitor.1:
1911 2003-05-03 Colin Walters <walters@verbum.org>
1913 * dbus/dbus-sysdeps.c (fill_user_info): Test against
1914 DBUS_UID_UNSET to determine whether to do a uid lookup or not.
1916 * Makefile.am: Update to use new .pc versioning scheme.
1918 2003-05-02 Havoc Pennington <hp@redhat.com>
1920 * bus/system.conf.in: allow send/receive to/from message bus
1923 2003-04-30 Havoc Pennington <hp@redhat.com>
1925 * configure.in: print a note when building with unit tests and
1928 2003-04-30 Havoc Pennington <hp@redhat.com>
1930 * Makefile.am: add a check-local that complains if you didn't
1931 configure with --enable-tests
1933 2003-04-29 Havoc Pennington <hp@redhat.com>
1935 * glib/dbus-gmain.c: docs cleanups
1937 * dbus/dbus-types.h: add docs on int64 types
1939 * dbus/dbus-memory.c: fix docs to avoid putting private API in
1940 public API docs section
1942 2003-04-29 Havoc Pennington <hp@redhat.com>
1944 * dbus-1.pc.in, dbus-glib-1.pc.in: rename these from
1945 dbus-1.0.pc.in, dbus-glib-1.0.pc.in. As these change with the
1946 parallel install API version, not with the D-BUS package version.
1948 * HACKING: move some of README over here
1950 * README: updates, and document API/ABI policy
1952 * configure.in: reindentation
1954 2003-04-29 Havoc Pennington <hp@redhat.com>
1956 * dbus/dbus.h: add "you have to define DBUS_API_SUBJECT_TO_CHANGE
1957 to use this library" to be sure people have the right
1960 2003-04-28 Havoc Pennington <hp@redhat.com>
1962 * configure.in: add --enable-docs which by default is auto yes if
1963 doxygen and db2html found, no otherwise; but can be forced on/off
1965 * doc/Makefile.am: conditionalize whether to build docs on
1968 2003-04-28 Havoc Pennington <hp@redhat.com>
1970 * configure.in: 0.10
1974 * bus/system.conf.in: add <includedir>system.d</includedir>
1976 * dbus/dbus-userdb.c (_dbus_user_database_lookup): fix bug when
1977 username was provided but not uid
1979 * bus/config-parser.c (struct BusConfigParser): keep track of
1980 whether the parser is toplevel or was included; change some
1981 of the error handling if it's included.
1983 2003-04-27 Havoc Pennington <hp@pobox.com>
1987 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
1988 report correct status if we finish processing authentication
1989 inside this function.
1991 * bus/activation.c (try_send_activation_failure): use
1992 bus_transaction_send_error_reply
1994 * bus/connection.c (bus_connection_get_groups): return an error
1995 explaining the problem
1997 * bus/bus.c (bus_context_check_security_policy): implement
1998 restriction here that inactive connections can only send the
1999 hello message. Also, allow bus driver to send anything to
2002 * bus/connection.c (bus_connection_complete): create the
2003 BusClientPolicy here instead of on-demand.
2004 (bus_connection_get_policy): don't return an error
2006 * dbus/dbus-message.c (dbus_message_new_error_reply): allow NULL
2007 sender field in message being replied to
2009 * bus/bus.c (bus_context_check_security_policy): fix silly typo
2010 causing it to return FALSE always
2012 * bus/policy.c (bus_client_policy_check_can_send): fix bug where
2013 we checked sender rather than destination
2015 2003-04-25 Havoc Pennington <hp@redhat.com>
2017 test suite is slightly hosed at the moment, will fix soon
2019 * bus/connection.c (bus_connections_expire_incomplete): fix to
2020 properly disable the timeout when required
2021 (bus_connection_set_name): check whether we can remove incomplete
2022 connections timeout after we complete each connection.
2024 * dbus/dbus-mainloop.c (check_timeout): fix this up a bit,
2025 probably still broken.
2027 * bus/services.c (bus_registry_acquire_service): implement max
2028 number of services owned, and honor allow/deny rules on which
2029 services a connection can own.
2031 * bus/connection.c (bus_connection_get_policy): report errors here
2033 * bus/activation.c: implement limit on number of pending
2036 2003-04-25 Havoc Pennington <hp@redhat.com>
2038 * dbus/dbus-transport.c (_dbus_transport_get_unix_user): fix bug
2039 where we used >= 0 instead of != DBUS_UID_UNSET.
2041 2003-04-25 Havoc Pennington <hp@redhat.com>
2043 * glib/dbus-gmain.c (remove_watch): fix for a crash when watches
2044 were toggled without add/remove, fix from Anders Gustafsson
2046 2003-04-24 Havoc Pennington <hp@redhat.com>
2048 * test/data/valid-config-files/basic.conf: add <limit> tags to
2051 * bus/config-parser.h, bus/config-parser.c, bus/bus.c: Implement
2052 <limit> tag in configuration file.
2054 2003-04-24 Havoc Pennington <hp@redhat.com>
2056 * bus/dispatch.c: somehow missed some name_is
2058 * dbus/dbus-timeout.c (_dbus_timeout_set_enabled)
2059 (_dbus_timeout_set_interval): new
2061 * bus/connection.c (bus_connections_setup_connection): record time
2062 when each connection is first set up, and expire them after the
2063 auth timeout passes.
2065 2003-04-24 Havoc Pennington <hp@redhat.com>
2067 * dbus/dbus-message.c (dbus_message_name_is): rename
2068 (dbus_message_service_is): rename
2069 (dbus_message_sender_is): rename
2070 (dbus_message_get_service): rename
2072 2003-04-24 Havoc Pennington <hp@redhat.com>
2074 * configure.in: add --enable-checks
2076 * dbus/dbus-message.c (dbus_message_new): reverse name/service arguments
2078 * dbus/dbus-connection.c (dbus_connection_preallocate_send): fix
2079 to use thread locks.
2080 (_dbus_connection_handler_destroyed_locked): move some private
2081 functions into proper docs group
2083 * dbus/dbus-internals.h: add _dbus_return_if_fail,
2084 _dbus_return_val_if_fail
2086 Throughout: use dbus_return_if_fail
2088 2003-04-23 James Willcox <jwillcox@gnome.org>
2091 * glib/dbus-gmain.c: (add_timeout), (wakeup_main), (create_source),
2092 (dbus_connection_setup_with_g_main),
2093 (dbus_server_setup_with_g_main):
2094 * glib/test-dbus-glib.c: (main):
2095 * glib/test-thread-client.c: (main):
2096 * glib/test-thread-server.c: (new_connection_callback), (main):
2097 * tools/dbus-monitor.c: (main):
2099 Added a GMainContext argument to dbus_connection_setup_with_g_main()
2100 and dbus_server_setup_with_g_main().
2102 2003-04-20 Havoc Pennington <hp@pobox.com>
2104 * doc/dbus-specification.sgml: document the restrictions on
2105 message and service names
2107 2003-04-22 Havoc Pennington <hp@redhat.com>
2109 * dbus/dbus-message.c, dbus/dbus-marshal.c: add 64-bit integer
2110 support, and do some code cleanups to share more code and
2111 speed up array marshal/demarshal.
2113 * dbus-1.0.pc.in (Cflags): put libdir include file in cflags
2115 * configure.in: generate dbus-arch-deps.h
2117 * dbus/dbus-protocol.h (DBUS_TYPE_INT64, DBUS_TYPE_UINT64): add
2120 2003-04-22 Havoc Pennington <hp@redhat.com>
2122 * test/data/valid-messages/opposite-endian.message: fix test
2123 to use proper type for rply field
2125 * test/data/invalid-messages: add tests for below validation
2127 * dbus/dbus-message.c (decode_header_data): validate field types,
2128 and validate that named fields are valid names
2129 (decode_name_field): consider messages in the
2130 org.freedesktop.Local. namespace to be invalid.
2132 * dbus/dbus-string.c (_dbus_string_validate_name): new
2134 2003-04-19 Havoc Pennington <hp@pobox.com>
2136 * bus/driver.c (bus_driver_handle_hello): check limits and
2137 return an error if they are exceeded.
2139 * bus/connection.c: maintain separate lists of active and inactive
2140 connections, and a count of each. Maintain count of completed
2141 connections per user. Implement code to check connection limits.
2143 * dbus/dbus-list.c (_dbus_list_unlink): export
2145 * bus/bus.c (bus_context_check_security_policy): enforce a maximum
2146 number of bytes in the message queue for a connection
2148 2003-04-18 Havoc Pennington <hp@pobox.com>
2150 * dbus/dbus-auth.c (record_mechanisms): memleak fixes
2152 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): fix some
2155 * dbus/dbus-keyring.c (add_new_key): fix a memleak, and
2156 on realloc be sure to update the pointer in the keyring
2158 * dbus/dbus-string.c (_dbus_string_zero): compensate for align
2159 offset to avoid writing to unallocated memory
2161 * dbus/dbus-auth.c (process_rejected): return FALSE if we fail to
2162 try the next mechanism, so we properly handle OOM
2164 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): fix double-free
2166 (_dbus_keyring_new): fix OOM bug
2167 (_dbus_keyring_new_homedir): always set error; impose a maximum
2168 number of keys we'll load from the file, mostly to speed up the
2169 test suite and make its OOM checks more useful, but also for
2172 * dbus/dbus-auth.c (process_error_server): reject authentication
2173 if we get an error from the client
2174 (process_cancel): on cancel, send REJECTED, per the spec
2175 (process_error_client): send CANCEL if we get an error from the
2178 2003-04-18 Havoc Pennington <hp@pobox.com>
2180 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix UMR in verbose
2183 * dbus/dbus-auth.c (handle_client_data_cookie_sha1_mech): fix OOM
2186 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): only whine
2187 about DBUS_TEST_HOMEDIR once
2189 * bus/Makefile.am (TESTS_ENVIRONMENT): put DBUS_TEST_HOMEDIR in
2192 * bus/dispatch.c (bus_dispatch_sha1_test): actually load sha1
2193 config file so we test the right thing
2195 Throughout: assorted docs improvements
2197 2003-04-18 Havoc Pennington <hp@pobox.com>
2199 * glib/dbus-gmain.c: adapt to watch changes
2201 * bus/bus.c, bus/activation.c, etc.: adjust to watch changes
2203 * dbus/dbus-server.h: remove dbus_server_handle_watch
2205 * dbus/dbus-connection.h: remove dbus_connection_handle_watch
2207 * dbus/dbus-watch.c (dbus_watch_handle): change DBusWatch to work
2208 like DBusTimeout, so we don't need dbus_connection_handle_watch
2211 2003-04-17 Havoc Pennington <hp@redhat.com>
2213 * dbus/dbus-userdb.c, dbus/dbus-sysdeps.c: redo all the passwd
2214 database usage so it all goes via the DBusUserDatabase cache.
2216 2003-04-17 Havoc Pennington <hp@redhat.com>
2218 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix logic so that if
2219 there was an OOM watch we skipped, we always return TRUE so we
2220 iterate again to have a look at it again. Fixes test suite hang.
2221 Code rearrangement also lets us lose some memset and only iterate
2222 over callbacks once.
2224 * bus/driver.c (bus_driver_handle_message): sense of test for
2227 2003-04-16 Havoc Pennington <hp@pobox.com>
2229 * doc/dbus-specification.sgml: make spec say serials are unsigned
2231 * dbus/dbus-message.h: change message serials to unsigned
2233 * dbus/dbus-connection.c: adapt to message serials being unsigned
2235 2003-04-15 Havoc Pennington <hp@pobox.com>
2237 * bus/bus.c: create and keep around a shared DBusUserDatabase
2240 * bus/connection.c (bus_connection_get_groups): don't cache
2241 groups for user in the connection object, since user database
2242 object now does that.
2244 2003-04-16 Havoc Pennington <hp@redhat.com>
2246 * dbus/dbus-message.c (_dbus_message_add_size_counter): keep a
2247 list of size counters
2248 (_dbus_message_loader_putback_message_link): put back a popped link
2250 * dbus/dbus-connection.c
2251 (dbus_connection_set_max_live_messages_size): rename
2253 (dbus_connection_get_outgoing_size): get size of outgoing
2255 (_dbus_connection_set_connection_counter): remove this cruft
2257 2003-04-14 Havoc Pennington <hp@redhat.com>
2259 * dbus/dbus-userdb.c: user database abstraction, mostly to get
2260 caching, but at some point we might want to be able to use a
2263 * bus/dispatch.c (bus_dispatch_sha1_test): add a test that uses
2264 SHA1 conf file to test the sha1 auth mechanism, since the regular
2265 test always uses EXTERNAL when available.
2268 test/data/valid-config-files/debug-allow-all-sha1.conf.in:
2269 add conf file that requires use of sha1 auth
2271 2003-04-13 Havoc Pennington <hp@pobox.com>
2273 * tools/dbus-send.c, tools/dbus-monitor.c: two utility programs
2274 from Philip Blundell to send messages and monitor them.
2276 2003-04-13 Havoc Pennington <hp@pobox.com>
2278 * dbus/dbus-mainloop.c: fix some reentrancy issues by refcounting
2281 * test/data/valid-config-files/debug-allow-all.conf.in: allow all
2284 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
2285 fix to only recover unused bytes if we're already authenticated
2286 (_dbus_transport_get_is_authenticated): fix to still mark us
2287 authenticated if there are unused bytes.
2289 * bus/dispatch.c: implement security policy checking
2291 * bus/connection.c (bus_transaction_send_from_driver): new
2293 * bus/bus.c (bus_context_check_security_policy): new
2295 * bus/dispatch.c (send_service_nonexistent_error): delete this,
2296 now we just set the DBusError and it gets converted to an error
2299 * bus/connection.c (allow_user_function): enable code using actual
2300 data from the config file
2302 * bus/policy.c (list_allows_user): handle wildcard rules for
2303 user/group connection perms
2305 2003-04-13 Havoc Pennington <hp@pobox.com>
2307 * bus/config-parser.c: Load up the BusPolicy and BusPolicyRules
2309 * dbus/dbus-sysdeps.c (_dbus_get_user_id): new function
2311 * bus/policy.c (bus_policy_append_mandatory_rule)
2312 (bus_policy_append_default_rule, bus_policy_append_user_rule)
2313 (bus_policy_append_group_rule): new functions
2315 2003-04-12 Havoc Pennington <hp@pobox.com>
2317 * bus/config-parser.c (bus_config_parser_new): fix a memleak
2319 * dbus/dbus-sysdeps.c: change DBusCredentials to use longs for
2320 the pid/gid/uid, just for paranoia.
2322 * test/break-loader.c (randomly_do_n_things): find a byte
2323 containing a type code, and randomly change it to a different
2326 2003-04-12 Havoc Pennington <hp@pobox.com>
2328 * bus/policy.h: change BusPolicy to be the thing from the config
2329 file, and rename old BusPolicy to BusClientPolicy
2331 * bus/bus.c, bus/connection.c, bus/config-parser.c: change to
2332 match change in how policy works
2334 * dbus/dbus-internals.h: mark assert_not_reached as
2335 __attribute((noreturn))__
2337 2003-04-11 Havoc Pennington <hp@redhat.com>
2339 * doc/dbus-specification.sgml: fix a spot with the wrong name for
2340 the broadcast service. Use boolean return for ServiceExists.
2342 2003-04-11 Havoc Pennington <hp@redhat.com>
2344 * configure.in: add another directory to look for qt in.
2346 2003-04-11 Havoc Pennington <hp@redhat.com>
2348 * AUTHORS: add Colin Walters
2350 2003-04-11 Havoc Pennington <hp@redhat.com>
2356 2003-04-11 Havoc Pennington <hp@redhat.com>
2358 * bus/messagebus.in: remove pid file when stopping the
2359 message bus, since the bus won't have privileges to remove it
2362 2003-04-11 Havoc Pennington <hp@redhat.com>
2364 * bus/bus.c (bus_context_new): move credentials change after
2367 2003-04-11 Havoc Pennington <hp@pobox.com>
2369 * test/decode-gcov.c: add "below average functions" to the
2370 coverage report, and change how some of the code works.
2372 * bus/test-main.c: bracket some stuff in DBUS_BUILD_TESTS so it's
2373 not in the coverage stats.
2375 * test/test-service.c (main): use _dbus_verbose not fprintf in a
2376 couple places so running the test suite doesn't result in megaspam.
2378 2003-04-11 Havoc Pennington <hp@pobox.com>
2380 * bus/dispatch.c (check_existent_service_activation): accept a no
2381 memory error in a place we didn't before
2383 * bus/test.c (bus_test_run_everything): remove hacky "do it twice
2384 in case the first one failed," since the test suite is less
2387 2003-04-10 Havoc Pennington <hp@pobox.com>
2389 * bus/dispatch.c (check_segfault_service_activation): add test
2390 for launching an executable that just crashes.
2392 * test/test-segfault.c (main): try setting coredumpsize to 0 so we
2393 don't leave a million cores. We'll see how portable this is.
2395 2003-04-10 Havoc Pennington <hp@pobox.com>
2397 * dbus/dbus-spawn.c (_dbus_spawn_async_with_babysitter): move all
2398 the possible parent failures before we fork, so that we don't
2399 fail to create a babysitter after creating the child.
2401 * bus/activation.c (bus_activation_activate_service): kill child
2402 if we don't successfully complete the activation.
2404 2003-04-10 Havoc Pennington <hp@redhat.com>
2406 * dbus/dbus-connection.c (dbus_connection_flush): don't spin on
2407 the connection if it's disconnected
2409 * bus/activation.c (bus_activation_service_created): use new
2410 transaction features to roll back removal of pending activation if
2411 we don't successfully create the service after all. Don't remove
2412 pending activation if the function fails.
2414 * dbus/dbus-list.c (_dbus_list_insert_before_link)
2415 (_dbus_list_insert_after_link): new code to facilitate
2418 * dbus/dbus-hash.c (_dbus_hash_table_insert_string_preallocated):
2419 new functionality, so we can preallocate the ability to insert
2422 * bus/connection.c (bus_transaction_add_cancel_hook): new function
2423 allowing us to put custom hooks in a transaction to be used for
2424 cancelling said transaction
2426 * doc/dbus-specification.sgml: add some discussion of secondary
2427 service owners, and disallow zero-length service names
2429 * bus/services.c (bus_registry_acquire_service): new function,
2430 splits out part of bus_driver_handle_acquire_service() and fixes
2431 a bug where we didn't remove the service doing the acquiring
2432 from the secondary queue if we failed to remove the current owner
2433 from the front of the queue.
2435 2003-04-10 Alexander Larsson <alexl@redhat.com>
2437 * doc/dbus-specification.sgml:
2438 s/org.freedesktop.Broadcast/org.freedesktop.DBus.Broadcast/
2440 2003-04-10 Alexander Larsson <alexl@redhat.com>
2445 Added files to cvsignore
2447 * dbus/dbus-message.h:
2448 * dbus/dbus-message.c: (dbus_message_iter_get_named):
2449 Make get_named() take two out argument and return a boolean.
2450 (dbus_message_iter_get_args_valist):
2451 Update usage of get_named().
2452 (dbus_message_iter_append_byte):
2454 (dbus_message_iter_append_named)
2456 (message_iter_test), (check_message_handling_type), (_dbus_message_test):
2459 2003-04-10 Alexander Larsson <alexl@redhat.com>
2461 * dbus/dbus-marshal.[ch]:
2462 Add array_type_pos argument to _dbus_marshal_validate_arg.
2463 Let you pass a NULL end_pos to _dbus_marshal_validate_type.
2465 * dbus/dbus-message.[ch]:
2466 Multi-dimensional arrays have full type specification in the
2467 outermost array. Iter code re-arranged to handle this.
2468 Added some more iter tests.
2470 * doc/dbus-specification.sgml:
2473 Update new array encoding description.
2474 Correct DBUS_SERVICE_FLAGS_REPLACE_EXISTING description.
2476 * test/data/invalid-messages/array-with-mixed-types.message:
2477 * test/data/valid-messages/array-of-array-of-uint32.message:
2478 Change to the new array format.
2480 * test/data/invalid-messages/too-short-dict.message:
2483 * test/data/valid-messages/recursive-types.message:
2484 Fix up and extend test.
2486 2003-04-10 Havoc Pennington <hp@pobox.com>
2488 * bus/dispatch.c: lots of fixes
2490 * dbus/dbus-mainloop.c (_dbus_loop_dispatch): export
2491 (_dbus_loop_iterate): remove old "quit if no callbacks" code,
2492 that was crack, broke the test service.
2494 * dbus/dbus-transport.c (_dbus_transport_open): fix error
2495 handling to avoid piling up errors if we get a failure on the
2498 * dbus/dbus-internals.c (_dbus_real_assert_not_reached): include
2499 pid in assertion failures.
2501 * dbus/dbus-mainloop.c (_dbus_loop_iterate): use static arrays up
2502 to some fixed size of file descriptor array. Don't return TRUE
2503 anytime a timeout exists, that led to lots of busy loop silliness
2506 2003-04-09 Havoc Pennington <hp@redhat.com>
2508 * dbus/dbus-mainloop.c (check_timeout): fix timeouts, I thought
2509 I'd checked this in earlier but hadn't.
2511 2003-04-09 Havoc Pennington <hp@redhat.com>
2513 * bus/dispatch.c (bus_dispatch_test): get a bit further through
2514 the activation test (man this is getting old!)
2516 2003-04-09 Havoc Pennington <hp@redhat.com>
2518 * test/test-utils.c: use dispatch status function to fix this up
2520 * bus/connection.c (connection_watch_callback): don't dispatch
2522 (connection_timeout_callback): don't dispatch from here
2523 (bus_connections_setup_connection): set the dispatch status function
2524 (bus_connection_disconnected): unset it
2526 * dbus/dbus-mainloop.c (_dbus_loop_queue_dispatch): new function
2527 used to add a connection to be dispatched
2528 (_dbus_loop_iterate): do the dispatching at the end of each
2531 * dbus/dbus-connection.c
2532 (dbus_connection_set_dispatch_status_function): new function
2533 allowing us to fix up main loop usage
2534 (_dbus_connection_last_unref): free all the various function
2536 (dbus_connection_dispatch): call the DispatchStatusFunction
2537 whenever this function returns
2538 (dbus_connection_handle_watch): call DispatchStatusFunction
2539 (dbus_connection_send_with_reply_and_block): call DispatchStatusFunction
2540 (reply_handler_timeout): call DispatchStatusFunction
2541 (dbus_connection_flush): call DispatchStatusFunction
2543 2003-04-09 Havoc Pennington <hp@redhat.com>
2545 * dbus/dbus-bus.c (dbus_bus_register): fix up error handling and
2548 * bus/dispatch.c (check_service_activated): fix bug in test
2550 * dbus/dbus-mainloop.c (check_timeout): fix this up
2552 * dbus/dbus-internals.c (_dbus_verbose_real): include PID in
2553 verbose output so we can sort out output from different processes,
2554 e.g. in the activation case.
2556 2003-04-08 Colin Walters <walters@gnu.org>
2558 * bus/bus.c (struct BusContext) [pidfile]: New member, to store
2560 (bus_context_new): Set it.
2561 (bus_context_unref): Use it to delete the pid file.
2563 2003-04-08 Havoc Pennington <hp@redhat.com>
2565 * test/data/invalid-messages/array-with-mixed-types.message:
2566 regression test that fails for the moment
2568 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): reorder
2569 tests for convenience
2571 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): don't allow
2572 array of nil, it broke things.
2574 * test/data/invalid-messages/array-of-nil.message: regression test
2576 * test/data/valid-messages/array-of-array-of-uint32.message:
2577 happened to write this so added it to suite
2579 2003-04-08 Havoc Pennington <hp@redhat.com>
2581 * bus/driver.c (bus_driver_handle_acquire_service): init
2582 retval/reply before checking name
2584 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add a
2585 recursion depth argument
2587 * dbus/dbus-message.h (struct DBusMessageIter): put some padding
2588 in the public struct for future extension
2590 * dbus/dbus-message-builder.c (_dbus_message_data_load): fix
2593 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix a verbose
2596 * doc/dbus-specification.sgml: fix typo
2598 2003-04-08 Alexander Larsson <alexl@redhat.com>
2600 Implemented recursive types, named types and new-style iters
2603 * glib/test-thread-client.c: (thread_func):
2604 * glib/test-thread-server.c: (handle_test_message):
2605 * test/test-service.c: (handle_echo):
2614 * dbus/dbus-internals.c: (_dbus_type_to_string):
2615 Update for new types.
2617 * dbus/dbus-marshal.[ch]:
2618 Implement recursive types and the new marshalling format.
2619 Remove hardcoded dict marshalling.
2620 Marshal named types.
2622 * dbus/dbus-message-builder.c:
2624 Remove references to old types
2626 * dbus/dbus-message.[ch]:
2627 New non-refcounted iter API that supports recursive iters.
2628 Use iters for appending, including support for recursive
2630 Add byte and named type support.
2631 Update everything to new marshalling formats.
2632 Add tests for new API.
2634 * dbus/dbus-protocol.h:
2635 Remove old array types.
2636 Add types: BYTE, ARRAY, DICT, NAMED
2638 * dbus/dbus-string.c:
2639 * dbus/dbus-sysdeps.c:
2640 Make parse_double locale safe.
2642 * dbus/dbus-test-main.c:
2648 * doc/dbus-specification.sgml:
2651 * test/data/incomplete-messages/missing-body.message:
2652 * test/data/invalid-messages/bad-boolean.message:
2653 * test/data/invalid-messages/bad-boolean-array.message:
2654 * test/data/invalid-messages/boolean-array-length-too-long.message-raw:
2655 * test/data/invalid-messages/boolean-has-no-value.message-raw:
2656 * test/data/invalid-messages/too-short-dict.message:
2657 * test/data/valid-messages/dict-simple.message:
2658 * test/data/valid-messages/dict.message:
2659 * test/data/valid-messages/emptiness.message:
2660 * test/data/valid-messages/lots-of-arguments.message:
2661 * test/data/valid-messages/no-padding.message:
2662 * test/data/valid-messages/recursive-types.message:
2663 Add missing NAME fields
2664 Fix up dicts & arrays
2666 * test/data/invalid-messages/dict-with-nil-value.message:
2667 Removed, this is not invalid anymore.
2669 * test/data/valid-messages/recursive-types.message:
2670 Add new test for deeply recursive types.
2672 2003-04-07 Havoc Pennington <hp@pobox.com>
2674 * bus/driver.c (bus_driver_handle_acquire_service): return an
2675 error if you try to acquire a service that starts with ':'
2677 2003-04-07 Havoc Pennington <hp@redhat.com>
2679 * doc/dbus-specification.sgml: require that base service names
2680 start with ':' and that the base service is created/deleted
2681 as first and last things a connection does on the bus
2683 * bus/dispatch.c (check_existent_service_activation): lots more
2684 work on the activation test; it doesn't fully pass yet...
2686 * test/test-service.c (main): fix so we don't memleak the
2687 connection to the message bus
2688 (filter_func): accept a message asking us to exit
2690 2003-04-06 Havoc Pennington <hp@pobox.com>
2692 * qt/Makefile.am (dbusinclude_HEADERS): install dbus-qt.h,
2695 * configure.in: fixes to Qt detection from Colin Walters
2697 * doc/Makefile.am: Only remove generated docbook dirs if they
2698 exist, from Colin Walters
2700 * dbus/dbus-bus.c: change how we set well-known connections to
2701 NULL, so that it works if a single connection is stored in
2702 two well-known array slots.
2704 * test/Makefile.am: remove a lot of stuff that isn't immediately
2705 useful, it's in CVS history if we want it.
2707 * test/test-service.c: use dbus-mainloop instead of that
2710 2003-04-06 Havoc Pennington <hp@pobox.com>
2712 * dbus/Makefile.am: split lists of sources into stuff that goes in
2713 the library, util functions that go in the lib and are also used
2714 elsewhere, and util functions that are used in tests/daemon but
2715 don't go in the lib.
2717 * dbus/dbus-mainloop.h, dbus/dbus-mainloop.c: move bus/loop.[hc]
2718 here so it can be used in test binaries also
2720 2003-04-06 Havoc Pennington <hp@pobox.com>
2722 * dbus/dbus-sysdeps.c (_dbus_become_daemon): write the pidfile
2723 here in the parent process, so we can return an error if it
2724 fails. Also, move some of the code into the child so the parent
2725 is less hosed if we fail midway through.
2727 * bus/bus.c (bus_context_new): move pidfile detection further up
2728 in the function, before we start overwriting sockets and such.
2730 * bus/messagebus.in: adjust this a bit, not sure if it will work.
2732 * configure.in: add --with-system-pid-file and --with-system-socket
2734 2003-04-06 Colin Walters <walters@verbum.org>
2736 * configure.in (DBUS_SYSTEM_PID_FILE): New variable.
2738 * bus/system.conf.in: Declare a pidfile.
2740 * bus/bus.c (bus_context_new): Test for an existing pid file, and
2741 create one (if appropriate).
2743 * bus/config-parser.c (enum ElementType) [ELEMENT_PIDFILE]: New.
2744 (struct BusConfigParser) [pidfile]: New.
2745 (element_type_to_name, merge_included, start_busconfig_child)
2746 (bus_config_parser_end_element, bus_config_parser_content): Handle it.
2747 (bus_config_parser_unref): Free it.
2748 (bus_config_parser_get_pidfile): New function.
2750 * bus/config-parser.h (_dbus_write_pid_file): Prototype.
2752 * dbus/dbus-errors.h (DBUS_ERROR_PIDFILE_EXISTS): New error.
2754 * dbus/dbus-sysdeps.c (_dbus_write_pid_file): New function.
2756 * dbus/dbus-sysdeps.h: Prototype it.
2758 2003-04-06 Havoc Pennington <hp@pobox.com>
2760 * bus/bus.c (bus_context_new): print the address in here, rather
2761 than in main(), because we need to do it before forking the daemon
2763 * bus/dispatch.c (send_service_nonexistent_error): set the sender
2764 on the service nonexistent error
2766 * bus/driver.c (bus_driver_handle_acquire_service): set the
2767 sender on the AcquireService reply
2769 * test/data/valid-config-files/debug-allow-all.conf.in: Make test
2770 server also listen on a UNIX socket so services can connect to it.
2772 2003-04-06 Havoc Pennington <hp@pobox.com>
2774 * dbus/dbus-threads.c: Redo how the fake debug mutexes are done
2775 so it detects deadlocks and also we actually init threads when
2778 2003-04-06 Havoc Pennington <hp@pobox.com>
2780 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
2781 save the domain socket name, and unlink it when we disconnect the
2782 server. Means that at least when we exit normally, we won't leave
2783 a bunch of junk in /tmp
2785 * dbus/dbus-transport-unix.c
2786 (_dbus_transport_new_for_domain_socket): code cleanup (nicer
2787 memory management). (I was making a real change here but then
2790 2003-04-06 Havoc Pennington <hp@pobox.com>
2792 * bus/bus.c (bus_context_new): fix wrong handling of
2793 server_data_slot_unref() in the error case.
2795 * dbus/dbus-internals.h (_dbus_assert): change so it passes
2796 "(condition) != 0" to _dbus_real_assert so that
2797 "_dbus_assert (pointer)" doesn't cause a warning
2799 * bus/main.c (main): accept --print-address option to print out
2800 the message bus address
2802 * dbus/dbus-sysdeps.c (_dbus_generate_random_ascii): export this
2804 * dbus/dbus-transport.c (_dbus_transport_open): special error for
2805 "tmpdir" option to unix: address on client side
2807 * dbus/dbus-server.c (dbus_server_listen): handle "tmpdir" option
2810 * configure.in (TEST_SOCKET_DIR): locate a temporary directory
2811 we can use to create sockets in the test suite.
2813 * bus/main.c (signal_handler): on SIGTERM, exit the daemon
2814 cleanly. To be used for testing.
2816 * dbus/dbus-spawn.c (babysit): use _dbus_set_signal_handler()
2818 * dbus/dbus-sysdeps.c (_dbus_set_signal_handler): new
2820 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
2821 handle trying to call this when there's no servers active
2823 2003-04-05 Havoc Pennington <hp@pobox.com>
2829 2003-04-05 Havoc Pennington <hp@pobox.com>
2831 * bus/bus.c (setup_server): fix this so dbus-daemon-1 doesn't
2832 crash on startup. Need to get "try starting the daemon"
2833 in the test suite I guess. ;-)
2835 * dbus/dbus-server.h, dbus/dbus-server.c: remove the stuff that
2836 tracked the number of open connections; it's better done in
2837 application-specific code as you want it to span all servers etc.
2839 2003-04-05 Havoc Pennington <hp@pobox.com>
2841 * bus/Makefile.am (install-data-hook): add missing DESTDIR,
2842 patch from Colin Walters
2844 2003-04-05 Havoc Pennington <hp@pobox.com>
2846 * doc/config-file.txt (Elements): fix docs of <auth> to reflect
2847 reality; in fact multiple mechanisms are allowed.
2849 * dbus/dbus-internals.c (_dbus_real_assert)
2850 (_dbus_real_assert_not_reached): move guts of _dbus_assert() and
2851 _dbus_assert_not_reached() into functions, so that they don't show
2852 up in basic block counts for test coverage, and don't use up as
2853 much disk space. Does mean slower execution speed though, so
2854 assumes --disable-asserts is the normal production case.
2856 2003-04-05 Havoc Pennington <hp@pobox.com>
2858 * test/Makefile.am (dist-hook): also dist *.in files
2864 2003-04-05 Havoc Pennington <hp@pobox.com>
2866 * dbus/dbus-string.c: docs warning
2868 * dbus/dbus-spawn.c: missing docs
2870 * dbus/dbus-memory.c (struct ShutdownClosure): missing docs
2872 2003-04-05 Havoc Pennington <hp@pobox.com>
2874 * bus/loop.c (bus_loop_iterate): fix the timeout code, using
2877 * dbus/dbus-spawn.c (_dbus_babysitter_unref): set sitter_pid
2878 to -1 once we've reaped the babysitter
2879 (_dbus_babysitter_handle_watch): do as much work as we can, not
2882 * bus/activation.c: add code using DBusBabysitter so that we
2883 handle it when a service fails to start up properly.
2884 (bus_activation_service_created): don't remove the activation
2885 entries as we go, just let them get removed when we free the pending
2886 activation. Unref reply messages after sending them.
2888 2003-04-05 Havoc Pennington <hp@pobox.com>
2890 * test/decode-gcov.c (main): print per-directory stats in the report
2892 * Makefile.am (coverage-report.txt): don't include test/* in gcov stats
2894 2003-04-05 Havoc Pennington <hp@pobox.com>
2896 * Makefile.am (coverage-report.txt): add target "coverage-report.txt"
2898 * test/decode-gcov.c: hack up a little program to suck data
2899 out of gcov files. Yes this is sort of silly.
2901 * configure.in: define something in config.h and do an
2902 AM_CONDITIONAL when gcov is enabled
2904 2003-04-04 Havoc Pennington <hp@redhat.com>
2906 * dbus/dbus-spawn.c, dbus/dbus-spawn.h: Change dbus_spawn to
2907 return a "babysitter" object that is used to monitor the status of
2908 the spawned process and reap it when required.
2910 * test/test-segfault.c, test/test-exit.c,
2911 test/test-sleep-forever.c: binaries that do various lame things,
2912 used in the test suite.
2914 * dbus/dbus-sysdeps.c: kill _dbus_errno_to_string()
2916 2003-04-03 Havoc Pennington <hp@pobox.com>
2918 * dbus/dbus-spawn.c: Move dbus-spawn into a separate file
2919 in preparation for modifying it, dbus-sysdeps is getting
2922 2003-04-03 Havoc Pennington <hp@redhat.com>
2924 * bus/loop.h, bus/loop.c: make the mainloop an object so we can
2927 * bus/*.[hc]: adapt to mainloop change
2929 2003-04-03 Havoc Pennington <hp@redhat.com>
2931 * bus/activation.c (load_directory): fix up memleaks
2932 (bus_activation_entry_free): free the entry
2934 * dbus/dbus-bus.c (dbus_bus_acquire_service): return an error if
2935 we get one from the message bus; fix memleaks.
2937 * dbus/dbus-message.c (dbus_set_error_from_message): new function
2939 2003-04-03 Havoc Pennington <hp@pobox.com>
2941 * bus/config-parser.c (bus_config_parser_unref): free
2942 list of mechanisms, bug discovered by test suite enhancements
2943 (putting system.conf and session.conf into suite)
2945 * test/Makefile.am, test/test-service.c: add placeholder for a
2946 test service that we'll activate as part of test suite. Doesn't
2949 * dbus/dbus-sysdeps.c (_dbus_setenv): support unsetenv by
2950 setting NULL value, and use system malloc not dbus_malloc()
2951 when we have unavoidable memleakage.
2953 * dbus/dbus-bus.c (dbus_bus_get): fix bug where bus type of 0
2954 didn't work, and support DBUS_BUS_ACTIVATION.
2956 * bus/activation.c (child_setup): pass our well-known bus type to
2959 * bus/config-parser.c: support <type> to specify well-known type
2961 * doc/dbus-specification.sgml: document the env variables to
2962 locate well-known buses and find service activator
2964 2003-04-02 Havoc Pennington <hp@redhat.com>
2966 * test/Makefile.am (all-local): add a rule to copy tests to
2967 builddir, so we can have generated tests. Use this to remove the
2968 silly hack for testing system.conf and session.conf. Will use this
2969 shortly to generate .service files pointing to test binaries.
2971 2003-04-02 Havoc Pennington <hp@redhat.com>
2973 * dbus/dbus-string.c (set_length): fix a bug - we allocated max of
2974 current alloc and needed new length, not max of the doubled
2975 allocation and needed new length. Also, when building tests,
2976 don't do the double-allocation stuff, just realloc every time.
2978 2003-04-02 Havoc Pennington <hp@redhat.com>
2980 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): include filenames
2982 (_dbus_string_get_dirname): new
2983 (_dbus_sysdeps_test): new
2984 (_dbus_directory_open): include dirnames in error messages
2986 * bus/config-parser.c: interpret <include> and <includedir> and
2987 <servicedir> relative to config file location if the given
2988 filename is not absolute.
2990 * dbus/dbus-string.c (_dbus_string_find_byte_backward): new
2992 2003-04-02 Havoc Pennington <hp@redhat.com>
2994 * bus/connection.c (bus_transaction_send_error_reply): set sender
2995 service for the error, and unref the reply on success
2997 * bus/activation.c: convert to use BusTransaction so OOM can be
2999 (bus_activation_service_created): set sender of the message
3001 2003-04-01 Havoc Pennington <hp@redhat.com>
3003 * bus/config-parser.c, bus/bus.c: implement <servicedir> and
3004 <includedir> (at least mostly)
3006 * dbus/dbus-sysdeps.c (_dbus_change_identity): set the group ID
3007 first, then the user ID
3009 2003-04-01 Havoc Pennington <hp@pobox.com>
3011 * dbus/dbus-server.c (dbus_server_set_auth_mechanisms): new
3014 * dbus/dbus-auth.c (_dbus_auth_set_mechanisms): new
3016 * dbus/dbus-internals.c (_dbus_dup_string_array): new function
3018 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): chmod the
3019 socket 0777, and unlink any existing socket.
3021 * bus/bus.c (bus_context_new): change our UID/GID and fork if
3022 the configuration file so specifies; set up auth mechanism
3025 * bus/config-parser.c (bus_config_parser_content): add support
3026 for <fork> option and fill in code for <auth>
3028 * bus/system.conf.in: add <fork/> to default configuration,
3029 and limit auth mechanisms to EXTERNAL
3031 * doc/config-file.txt (Elements): add <fork>
3033 * dbus/dbus-sysdeps.c (_dbus_become_daemon): new function
3034 (_dbus_change_identity): new function
3036 2003-03-31 Havoc Pennington <hp@redhat.com>
3038 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket)
3039 (_dbus_listen_unix_socket): fix off-by-one error in null
3040 termination spotted by Nalin
3042 2003-03-31 Havoc Pennington <hp@redhat.com>
3044 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): allow setting
3045 DBUS_TEST_HOMEDIR when tests are enabled, so we can test without
3046 having a real home directory available.
3048 2003-03-31 Havoc Pennington <hp@redhat.com>
3050 * bus/Makefile.am (install-data-hook): create /var/run/dbus
3052 * bus/messagebus.in: add init script for Red Hat /etc/init.d
3054 * configure.in: add support for specifying a style of init script
3057 2003-03-31 Havoc Pennington <hp@redhat.com>
3059 Fix some annoying DBusString API and fix all affected code.
3061 * dbus/dbus-string.c (_dbus_string_init): get rid of annoying
3063 (_dbus_string_get_data): change to return string instead of using
3065 (_dbus_string_get_const_data): ditto
3066 (_dbus_string_get_data_len): ditto
3067 (_dbus_string_get_const_data_len): ditto
3069 2003-03-31 Havoc Pennington <hp@redhat.com>
3071 * bus/main.c (main): fix up the command line arguments to be nicer
3073 2003-03-31 Havoc Pennington <hp@redhat.com>
3075 * dbus/Makefile.am (INCLUDES): use EXPANDED_LOCALSTATEDIR to
3076 define DBUS_SYSTEM_BUS_PATH as we want to compile in the same
3077 final location that lands in the config file
3079 * bus/config-loader-expat.c (bus_config_load): fix type of
3082 * doc/TODO: remove TODO item for dbus_bus_get()
3084 * dbus/dbus-bus.c (bus_data_free): add missing lock/unlock
3086 2003-03-31 Havoc Pennington <hp@pobox.com>
3088 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_domain_socket)
3089 (_dbus_transport_new_for_tcp_socket): these didn't need the "server"
3090 argument since they are always client side
3092 * dbus/dbus-server.c (dbus_server_get_address): new function
3094 * bus/main.c (main): take the configuration file as an argument.
3096 * test/data/valid-config-files/debug-allow-all.conf: new file to
3097 use with dispatch.c tests for example
3099 * bus/test-main.c (main): require test data dir
3101 * bus/bus.c (bus_context_new): change this to take a
3102 configuration file name as argument
3104 * doc/config-file.txt (Elements): add <servicedir>
3106 * bus/system.conf, bus/session.conf: new files
3108 * dbus/dbus-bus.c (dbus_bus_get): look for system bus on
3109 well-known socket if none set
3111 * configure.in: create system.conf and session.conf
3113 2003-03-30 Havoc Pennington <hp@pobox.com>
3115 * bus/config-parser.c: hacking
3117 * dbus/dbus-memory.c: don't use DBusList for the list of stuff
3118 to shut down, since it could cause weirdness with the DBusList
3121 * dbus/dbus-list.c (_dbus_list_test): add tests for the
3122 link-oriented stack routines
3123 (alloc_link): free the mempool if the first alloc from it fails
3125 * dbus/dbus-mempool.c (struct DBusMemBlock): fix alignment issue
3127 * dbus/dbus-string.c (UNICODE_VALID): sync new version of this
3129 (_dbus_string_skip_white): new
3131 * doc/config-file.txt (Elements): add <includedir>
3133 2003-03-28 Havoc Pennington <hp@pobox.com>
3135 * dbus/dbus-string.c (_dbus_string_copy_data_len)
3136 (_dbus_string_copy_data): new functions
3138 2003-03-28 Anders Carlsson <andersca@codefactory.se>
3140 * dbus/dbus-bus.c: (bus_data_free), (dbus_bus_get):
3144 * dbus/dbus-memory.c:
3147 2003-03-28 Havoc Pennington <hp@pobox.com>
3149 * bus/test.c (bus_test_flush_bus): remove the sleep from here,
3150 I think it may have just been superstition. Not sure.
3152 * dbus/dbus-string.c (_dbus_string_base64_decode): catch some OOM
3153 failures that were not being handled.
3155 * dbus/dbus-auth.c (process_auth): fix a memleak in OOM handling
3157 * dbus/dbus-memory.c: add ability to set number of mallocs in a
3158 row that will fail on out-of-memory.
3160 * dbus/dbus-internals.c (_dbus_test_oom_handling): convenience
3161 function for testing out-of-memory handling.
3163 * bus/config-loader-expat.c (memsuite): don't wrap the dbus
3164 allocation functions, they do map exactly to the expat ones.
3166 2003-03-27 Havoc Pennington <hp@redhat.com>
3168 * bus/config-loader-libxml.c (bus_config_load): add another error
3171 2003-03-26 Anders Carlsson <andersca@codefactory.se>
3174 Add note about automatic service activation.
3176 * doc/dbus-specification.sgml:
3177 Rename the specification and clarify a few things.
3179 2003-03-26 Anders Carlsson <andersca@codefactory.se>
3182 * dbus/dbus-address.c:
3184 * dbus/dbus-marshal.c:
3185 * dbus/dbus-server-debug-pipe.c:
3186 * dbus/dbus-transport-unix.c:
3187 Fix documentation warnings.
3189 2003-03-26 Havoc Pennington <hp@pobox.com>
3191 * bus/test-main.c, dbus/dbus-test.c (main): check memleaks
3192 after every test so it's quick and easy to see which leaked, and
3193 so we test multiple dbus_shutdown() calls
3195 * configure.in: change configure.in XML stuff to also support
3198 * config-loader-libxml.c: some hacking
3200 * config-loader-expat.c: some hacking
3202 * config-parser.c: some hacking, plus tests
3204 2003-03-25 Havoc Pennington <hp@redhat.com>
3206 * throughout - add more _DBUS_ASSERT_ERROR_IS_CLEAR
3208 * configure.in: add --with-xml option to specify XML library,
3209 right now only libxml is supported.
3211 * bus/config-loader-libxml.c, config-parser.c: sync some minor
3212 nonworking code between home and work, still just stubs
3214 2003-03-24 Havoc Pennington <hp@redhat.com>
3216 * dbus/dbus-sysdeps.c (_dbus_set_fd_nonblocking): move to this
3219 * dbus/dbus-errors.c (dbus_set_error, dbus_set_error_const): allow
3220 NULL argument for "message" if the error is a well-known one,
3221 fill in a generic message in this case.
3223 * dbus/dbus-errors.h (DBusResultCode): Kill DBusResultCode in
3226 * bus/test.c (bus_test_flush_bus): add
3228 * bus/policy.c (bus_policy_test): test code stub
3230 2003-03-24 Havoc Pennington <hp@pobox.com>
3232 * bus/connection.c (bus_connections_setup_connection): set up
3233 the "can this user connect" function, but it always returns
3234 TRUE until we have a config file parser so we can have a config
3235 file that allows connections.
3237 2003-03-23 Havoc Pennington <hp@pobox.com>
3239 * dbus/dbus-threads.c (dbus_mutex_new, dbus_condvar_new): with
3240 DBUS_BUILD_TESTS, actually alloc/free a block of memory for
3241 the mutex, so we can check for proper memory management
3244 * dbus/dbus-dataslot.c: remove the mutex from
3245 DBusDataSlotAllocator and lock it manually when using it,
3246 to simplify fitting it into the global slots framework.
3248 * dbus/dbus-threads.c (init_static_locks): rework how we're
3249 handling global locks so they are easily shut down.
3251 * bus/policy.c (bus_policy_append_rule): fix
3253 * bus/test-main.c (main): check for memleaks
3255 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): make
3256 test suite check for memleaks
3258 * dbus/dbus-memory.c: add support in test mode for tracking
3259 number of outstanding blocks
3261 2003-03-23 Havoc Pennington <hp@pobox.com>
3263 * bus/policy.c, bus/bus.c, bus/connection.c: implement allow/deny
3266 * dbus/dbus-hash.h: add ULONG hash keys
3268 * dbus/dbus-sysdeps.c (_dbus_get_groups): new
3269 (_dbus_get_group_id): new function
3271 2003-03-20 Havoc Pennington <hp@redhat.com>
3273 * dbus/dbus-connection.c (dbus_connection_set_unix_user_function):
3275 (dbus_connection_get_unix_user): new function
3277 2003-03-20 Havoc Pennington <hp@pobox.com>
3279 * bus/connection.c (bus_connection_send_oom_error): assert that
3280 message has a sender
3281 (connection_execute_transaction): ditto
3282 (bus_connection_preallocate_oom_error): fix to set the sender, and
3283 set recipient to the destination service, not the bus driver
3285 * bus/policy.c: hacking
3287 * dbus/dbus-message.c (dbus_message_service_is): new function
3288 (dbus_message_sender_is): new
3290 2003-03-19 Havoc Pennington <hp@redhat.com>
3292 * bus/policy.c: start sketching code for policy restrictions on
3293 what connections can do.
3295 2003-03-18 Havoc Pennington <hp@redhat.com>
3297 * doc/TODO: some notes on high-level todo items. Little nitpick
3298 stuff is all in @todo, so no need to add it here.
3300 * doc/config-file.txt: some notes on how config file might look
3302 2003-03-18 Anders Carlsson <andersca@codefactory.se>
3308 2003-03-17 Havoc Pennington <hp@redhat.com>
3310 * dbus/dbus-internals.h: add gcc attributes so that
3311 our printf-style functions warn on bad arguments to
3314 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): fix printf
3317 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
3320 2003-03-17 Havoc Pennington <hp@redhat.com>
3322 * bus/test-main.c (main): make it print something as it runs
3323 so make check doesn't look stuck
3325 * doc/negotiation.txt, doc/dbus-sasl-profile.txt: remove
3326 from CVS, now obsolete
3328 2003-03-17 Anders Carlsson <andersca@codefactory.se>
3330 * bus/dispatch.c: (bus_dispatch):
3331 Refetch the service name since it may have been reallocated
3332 when dbus_message_set_sender was called.
3334 * dbus/dbus-sysdeps.c: (_dbus_accept):
3335 Add address and address length variables and use them to stop
3336 valgrind from complaining.
3338 2003-03-17 Havoc Pennington <hp@pobox.com>
3340 All tests pass, no memleaks, no valgrind complaints.
3342 * bus/test.c: refcount handler_slot
3344 * bus/connection.c (bus_connections_new): refcount
3345 connection_data_slot
3347 * dbus/dbus-auth-script.c (_dbus_auth_script_run): delete unused
3348 bytes so that auth scripts pass.
3350 * bus/dispatch.c: init message_handler_slot so it gets allocated
3353 * bus/dispatch.c (message_handler_slot_ref): fix memleak
3355 * dbus/dbus-server-debug-pipe.c (_dbus_server_debug_pipe_new):
3356 dealloc server_pipe_hash when no longer used for benefit of
3359 * dbus/dbus-auth.c (process_command): memleak fix
3361 * bus/dispatch.c (check_hello_message): memleak fix
3363 2003-03-16 Havoc Pennington <hp@pobox.com>
3365 * dbus/dbus-bus.c (ensure_bus_data): fix double-unref of the data slot
3367 2003-03-17 Anders Carlsson <andersca@codefactory.se>
3369 * bus/activation.c (bus_activation_activate_service): Append
3370 the pending activation entry to the list of pending activations.
3372 2003-03-16 Havoc Pennington <hp@pobox.com>
3374 * bus/dispatch.c (bus_dispatch_test): remove double-unrefs of
3377 * dbus/dbus-address.c (create_entry): fix OOM handling when
3378 failing to alloc entry->method
3380 2003-03-16 Havoc Pennington <hp@pobox.com>
3382 * dbus/dbus-watch.c (_dbus_watch_new): handle failure to malloc
3385 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
3386 add some missing dbus_set_result
3388 * bus/dispatch.c (bus_dispatch_add_connection): handle failure to
3389 alloc the DBusMessageHandler
3391 * dbus/dbus-transport.c (_dbus_transport_disconnect): don't ref
3392 the transport here, since we call this from the finalizer; it
3393 resulted in a double-finalize.
3395 * dbus/dbus-transport.c (_dbus_transport_disconnect): fix a bug
3396 where we tried to use transport->connection that was NULL,
3397 happened when transport was disconnected early on due to OOM
3399 * bus/*.c: adapt to handle OOM for watches/timeouts
3401 * dbus/dbus-transport-unix.c: port to handle OOM during
3404 * dbus/dbus-auth.c (_dbus_auth_get_unused_bytes): return a
3405 reference to unused bytes instead of a copy
3407 * dbus/dbus-server.c (dbus_server_handle_watch): return FALSE for
3410 * dbus/dbus-connection.c (dbus_connection_handle_watch): return
3413 * dbus/dbus-timeout.c (dbus_timeout_handle): return FALSE for out
3416 2003-03-16 Anders Carlsson <andersca@codefactory.se>
3418 * doc/dbus-specification.sgml:
3419 Document reply message for ActivateService.
3421 2003-03-16 Anders Carlsson <andersca@codefactory.se>
3423 * bus/activation.c: (bus_pending_activation_entry_free),
3424 (bus_pending_activation_free), (bus_activation_new),
3425 (bus_activation_unref), (bus_activation_service_created),
3426 (bus_activation_activate_service):
3428 * bus/bus.c: (bus_context_new):
3429 * bus/desktop-file.c: (new_section):
3430 * bus/driver.c: (bus_driver_send_service_deleted),
3431 (bus_driver_handle_activate_service):
3432 * bus/services.c: (bus_registry_new), (bus_registry_ensure):
3434 * dbus/dbus-connection.c:
3435 (dbus_connection_send_with_reply_and_block):
3436 * dbus/dbus-message.c: (dbus_message_append_args_valist):
3437 * dbus/dbus-protocol.h:
3438 Make activation work better. Now pending activations will be queued
3439 and the daemon won't try to activate services that are already registered.
3441 2003-03-16 Havoc Pennington <hp@pobox.com>
3443 * dbus/dbus-bus.c (ensure_bus_data): handle failure to set
3446 * dbus/dbus-memory.c (_dbus_initialize_malloc_debug): support
3447 DBUS_MALLOC_BACKTRACES to print trace when failing an alloc
3449 2003-03-16 Havoc Pennington <hp@pobox.com>
3451 * dbus/dbus-string.c (_dbus_string_validate_utf8): oops, unbreak
3452 this. always run the test suite before commit...
3454 * bus/*: adapt to DBusConnection API changes
3456 * glib/dbus-gmain.c: adapt to DBusConnection API changes,
3457 requires renaming stuff to avoid dbus_connection_dispatch name
3460 * dbus/dbus-transport.c (_dbus_transport_queue_messages): new
3463 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
3464 separate from _dbus_message_loader_return_buffer()
3466 * dbus/dbus-connection.c (dbus_connection_get_n_messages): remove
3467 this, because it's now always broken to use; the number of
3468 messages in queue vs. the number still buffered by the message
3469 loader is undefined/meaningless. Should use
3470 dbus_connection_get_dispatch_state().
3471 (dbus_connection_dispatch): rename from
3472 dbus_connection_dispatch_message
3474 2003-03-16 Havoc Pennington <hp@pobox.com>
3476 * dbus/dbus-string.c (_dbus_string_validate_utf8): copy in a real
3479 2003-03-16 Anders Carlsson <andersca@codefactory.se>
3481 * dbus/dbus-connection.c:
3482 (dbus_connection_send_with_reply_and_block):
3483 Decrease connection->n_incoming when removing an entry
3485 * dbus/dbus-dict.c: (dbus_dict_entry_free),
3486 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
3487 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
3488 (dbus_dict_set_byte_array), (dbus_dict_set_string_array),
3489 (dbus_dict_get_boolean_array), (dbus_dict_get_double_array),
3490 (dbus_dict_get_byte_array):
3491 Handle NULL arrays and strings. Also add support for byte arrays.
3493 * dbus/dbus-marshal.c: (_dbus_marshal_byte_array),
3494 (_dbus_marshal_dict), (_dbus_demarshal_byte_array),
3495 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
3496 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
3497 (_dbus_demarshal_dict), (demarshal_and_validate_len),
3498 (_dbus_marshal_validate_arg), (_dbus_marshal_test):
3499 * dbus/dbus-marshal.h:
3500 Add support for marshalling and demarshalling empty arrays and strings.
3502 * dbus/dbus-message.c: (dbus_message_append_args_valist),
3503 (dbus_message_append_string_array),
3504 (dbus_message_iter_get_boolean),
3505 (dbus_message_iter_get_boolean_array),
3506 (dbus_message_iter_get_int32_array),
3507 (dbus_message_iter_get_uint32_array),
3508 (dbus_message_iter_get_double_array),
3509 (dbus_message_iter_get_byte_array),
3510 (dbus_message_iter_get_string_array), (dbus_message_iter_get_dict),
3511 (check_message_handling):
3512 Add support for getting empty arrays and dicts.
3514 * dbus/dbus-string.c: (_dbus_string_validate_utf8):
3515 Don't do any validation at all for now, that's better than just checking
3518 * test/data/valid-messages/emptiness.message:
3519 New test message with lots of empty arrays.
3521 2003-03-16 Havoc Pennington <hp@pobox.com>
3523 * dbus/dbus-connection.c
3524 (_dbus_connection_queue_received_message_link): new function that
3525 can't fail due to OOM
3527 * dbus/dbus-message.c (_dbus_message_loader_pop_message_link):
3528 new function pops a message together with a list link
3531 * dbus/dbus-transport-unix.c (queue_messages): use new link-based
3532 message queuing functions to avoid needing to alloc memory
3534 2003-03-16 Havoc Pennington <hp@pobox.com>
3536 Oops - test code was only testing failure of around 30 of the
3537 mallocs in the test path, but it turns out there are 500+
3538 mallocs. I believe this was due to misguided linking setup such
3539 that there was one copy of dbus_malloc etc. in the daemon and one
3540 in the shared lib, and only daemon mallocs were tested. In any
3541 case, the test case now tests all 500+ mallocs, and doesn't pass
3542 yet, though there are lots of fixes in this patch.
3544 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
3545 this so that it doesn't need to allocate memory, since it
3546 has no way of indicating failure due to OOM (and would be
3547 annoying if it did).
3549 * dbus/dbus-list.c (_dbus_list_pop_first_link): new function
3551 * bus/Makefile.am: rearrange to create two self-contained
3552 libraries, to avoid having libraries with overlapping symbols.
3553 that was resulting in weirdness, e.g. I'm pretty sure there
3554 were two copies of global static variables.
3556 * dbus/dbus-internals.c: move the malloc debug stuff to
3559 * dbus/dbus-list.c (free_link): free list mempool if it becomes
3562 * dbus/dbus-memory.c (_dbus_disable_mem_pools): new function
3564 * dbus/dbus-address.c (dbus_parse_address): free list nodes
3567 * bus/dispatch.c (bus_dispatch_add_connection): free
3568 message_handler_slot when no longer using it, so
3569 memory leak checkers are happy for the test suite.
3571 * dbus/dbus-server-debug-pipe.c (debug_finalize): free server name
3573 * bus/bus.c (new_connection_callback): disconnect in here if
3574 bus_connections_setup_connection fails.
3576 * bus/connection.c (bus_connections_unref): fix to free the
3578 (bus_connections_setup_connection): if this fails, don't
3579 disconnect the connection, just be sure there are no side
3582 * dbus/dbus-string.c (undo_alignment): unbreak this
3584 * dbus/dbus-auth.c (_dbus_auth_unref): free some stuff we were
3586 (_dbus_auth_new): fix the order in which we free strings
3589 * bus/connection.c (bus_connection_disconnected): fix to
3590 not send ServiceDeleted multiple times in case of memory
3593 * dbus/dbus-bus.c (dbus_bus_get_base_service): new function to
3594 get the base service name
3595 (dbus_bus_register_client): don't return base service name,
3596 instead store it on the DBusConnection and have an accessor
3598 (dbus_bus_register_client): rename dbus_bus_register()
3600 * bus/dispatch.c (check_hello_message): verify that other
3601 connections on the bus also got the correct results, not
3602 just the one sending hello
3604 2003-03-15 Havoc Pennington <hp@pobox.com>
3606 Make it pass the Hello handling test including all OOM codepaths.
3607 Now to do other messages...
3609 * bus/services.c (bus_service_remove_owner): fix crash when
3610 removing owner from an empty list of owners
3611 (bus_registry_ensure): don't leave service in the list of
3612 a connection's owned services if we fail to put the service
3615 * bus/connection.c (bus_connection_preallocate_oom_error): set
3616 error flag on the OOM error.
3618 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
3619 handle _dbus_transport_set_connection failure
3621 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd): modify
3622 to create watches up front and simply enable/disable them as
3624 (unix_connection_set): this can now fail on OOM
3626 * dbus/dbus-timeout.c, dbus/dbus-watch.c: add concept
3627 of enabling/disabling a watch or timeout.
3629 * bus/loop.c (bus_loop_iterate): don't touch disabled
3632 * glib/dbus-gmain.c: adapt to enable/disable watches and timeouts
3634 2003-03-15 Havoc Pennington <hp@pobox.com>
3636 * bus/dispatch.c (bus_dispatch_test): OK, now finally actually
3637 write useful test code, after all that futzing around ;-)
3639 Test does not yet pass because we can't handle OOM in
3640 _dbus_transport_messages_pending (basically,
3641 dbus_connection_preallocate_send() does not prealloc the write
3642 watch). To fix this, I think we need to add new stuff to
3643 set_watch_functions, namely a SetEnabled function so we can alloc
3644 the watch earlier, then enable it later.
3646 * dbus/Makefile.am (libdbus_convenience_la_SOURCES): move
3647 dbus-memory.c to the convenience lib
3649 * bus/test.c: rename some static functions to keep them clearly
3650 distinct from stuff in connection.c. Handle client disconnection.
3652 2003-03-14 Havoc Pennington <hp@pobox.com>
3654 * bus/dispatch.c (bus_dispatch_test): do test using debug-pipe
3655 transport, tests more of the real codepath. Set up clients
3656 with bus_setup_debug_client.
3658 * bus/test.c (bus_setup_debug_client): function to set up debug
3659 "clients" on the main loop
3661 * dbus/dbus-transport.c (_dbus_transport_open): add debug-pipe
3664 * dbus/dbus-server.c (dbus_server_listen): add debug-pipe
3667 * dbus/dbus-server-debug.c: support a debug server based on pipes
3669 * dbus/dbus-sysdeps.c (_dbus_full_duplex_pipe): new function
3670 (_dbus_close): new function
3672 * configure.in: check for socketpair
3674 2003-03-14 Havoc Pennington <hp@redhat.com>
3676 * dbus/dbus-memory.c: add a "detect buffer overwrites on free"
3679 * dbus/dbus-transport-debug.c: rework this a good bit to be
3680 less complicated. hopefully still works.
3682 * dbus/dbus-server-debug.c (handle_new_client): remove timeout
3685 * glib/dbus-gmain.c (timeout_handler): don't remove timeout
3688 * dbus/dbus-message.c (dbus_message_copy): rename from
3689 dbus_message_new_from_message, fix it up to copy
3690 all the message fields, add test case
3692 * bus/dispatch.c (bus_dispatch_test): add some more test code,
3693 not quite passing yet
3695 2003-03-14 Havoc Pennington <hp@pobox.com>
3697 * bus/loop.c (bus_loop_iterate): add this so we can "run loop
3698 until no work remains" in test code. (the large diff here
3699 is just code movement, no actual changes)
3701 * dbus/dbus-server-debug.c (DEFAULT_INTERVAL): change interval to
3702 1, no point waiting around for test code.
3703 (_dbus_server_debug_accept_transport): unref the timeout
3704 after adding it (right?)
3706 * dbus/dbus-transport-debug.c (DEFAULT_INTERVAL): ditto
3708 2003-03-13 Havoc Pennington <hp@redhat.com>
3710 * dbus/dbus-timeout.c (_dbus_timeout_list_set_functions): handle
3713 * dbus/dbus-watch.c (_dbus_watch_list_set_functions): handle out
3716 * dbus/dbus-connection.h: Make AddWatchFunction and
3717 AddTimeoutFunction return a bool so they can fail on out-of-memory
3719 * bus/bus.c (bus_context_new): set up timeout handlers
3721 * bus/connection.c (bus_connections_setup_connection): set up
3724 * glib/dbus-gmain.c: adapt to the fact that set_functions stuff
3727 * bus/bus.c (bus_context_new): adapt to changes
3729 * bus/connection.c: adapt to changes
3731 * test/watch.c: adapt to DBusWatch changes
3733 * bus/dispatch.c (bus_dispatch_test): started adding this but
3736 2003-03-14 Anders Carlsson <andersca@codefactory.se>
3738 * bus/dispatch.c (send_service_nonexistent_error): Fix typo.
3740 2003-03-13 Havoc Pennington <hp@pobox.com>
3742 * bus/test.c, bus/test.h, bus/Makefile.am, bus/test-main.c:
3743 set up a test framework as for the library
3745 2003-03-12 Havoc Pennington <hp@pobox.com>
3747 Throughout: purge global variables, introduce BusActivation,
3748 BusConnections, BusRegistry, etc. objects instead.
3750 * bus/bus.h, bus/bus.c: introduce BusContext as a global
3753 * test/Makefile.am (TEST_BINARIES): disable bus-test for now,
3754 going to redo this a bit differently I think
3756 2003-03-12 Havoc Pennington <hp@redhat.com>
3758 Mega-patch that gets the message bus daemon initially handling
3759 out-of-memory. Work still needed. Also lots of random
3760 moving stuff to DBusError instead of ResultCode.
3762 * dbus/dbus-list.c (_dbus_list_length_is_one): new function
3764 * dbus/dbus-connection.c
3765 (dbus_connection_send_with_reply_and_block): use DBusError
3767 * dbus/dbus-bus.c: adapt to API changes, make it use DBusError not
3770 * dbus/dbus-connection.c (dbus_connection_send): drop the result
3771 code here, as the only failure possible is OOM.
3773 * bus/connection.c (bus_connection_disconnect):
3774 rename bus_connection_disconnected as it's a notification only
3776 * bus/driver.c (bus_driver_handle_acquire_service): don't free
3777 "name" on get_args failure, should be done by get_args;
3778 don't disconnect client for bad args, just return an error.
3779 (bus_driver_handle_service_exists): ditto
3781 * bus/services.c (bus_services_list): NULL-terminate returned array
3783 * bus/driver.c (bus_driver_send_service_lost)
3784 (bus_driver_send_service_acquired): send messages from driver to a
3785 specific client to the client's unique name, not to the broadcast
3788 * dbus/dbus-message.c (decode_header_data): reject messages that
3789 contain no name field
3790 (_dbus_message_get_client_serial): rename to
3791 dbus_message_get_serial and make public
3792 (_dbus_message_set_serial): rename from set_client_serial
3793 (_dbus_message_set_reply_serial): make public
3794 (_dbus_message_get_reply_serial): make public
3796 * bus/connection.c (bus_connection_foreach): allow stopping
3797 iteration by returning FALSE from foreach function.
3799 * dbus/dbus-connection.c (dbus_connection_send_preallocated)
3800 (dbus_connection_free_preallocated_send)
3801 (dbus_connection_preallocate_send): new API for sending a message
3802 without possibility of malloc failure.
3803 (dbus_connection_send_message): rename to just
3804 dbus_connection_send (and same for whole function family)
3806 * dbus/dbus-errors.c (dbus_error_free): make this reinit the error
3808 * dbus/dbus-sysdeps.c (_dbus_exit): new function
3810 * bus/activation.c: handle/return errors
3812 * dbus/dbus-errors.h: add more DBUS_ERROR #define
3814 * dbus/dbus-sysdeps.c (_dbus_directory_open) (_dbus_file_get_contents)
3815 (_dbus_directory_get_next_file): use DBusError instead of DBusResultCode
3816 (_dbus_result_from_errno): move to this file
3818 2003-03-10 Anders Carlsson <andersca@codefactory.se>
3820 * dbus/dbus-marshal.c:
3821 (_dbus_marshal_set_string):
3822 Take a length argument so we can marshal the correct string
3825 (_dbus_marshal_dict), (_dbus_demarshal_dict),
3826 (_dbus_marshal_get_arg_end_pos), (_dbus_marshal_validate_arg),
3827 (_dbus_marshal_test):
3828 * dbus/dbus-marshal.h:
3829 Add support for marshalling and demarshalling dicts.
3831 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
3832 Add support for TYPE DICT.
3834 * dbus/dbus-message.c: (set_string_field):
3835 Adjust header padding.
3837 (dbus_message_append_args_valist), (dbus_message_append_dict),
3838 (dbus_message_get_args_valist), (dbus_message_iter_get_arg_type),
3839 (dbus_message_iter_get_dict), (_dbus_message_loader_return_buffer),
3840 (check_message_handling), (check_have_valid_message):
3841 * dbus/dbus-message.h:
3842 Add functions for setting and getting dicts.
3844 * dbus/dbus-protocol.h:
3850 * doc/dbus-specification.sgml:
3851 Add information about how dicts are marshalled.
3853 * test/data/invalid-messages/dict-with-nil-value.message:
3854 * test/data/invalid-messages/too-short-dict.message:
3855 * test/data/valid-messages/dict-simple.message:
3856 * test/data/valid-messages/dict.message:
3857 Add sample messages containing dicts.
3859 2003-03-08 Anders Carlsson <andersca@codefactory.se>
3861 * dbus/dbus-dict.h: Add DBUS_END_DECLS.
3863 2003-03-07 Anders Carlsson <andersca@codefactory.se>
3866 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
3867 (dbus_dict_get_keys), (insert_entry), (dbus_dict_set_boolean),
3868 (dbus_dict_set_int32), (dbus_dict_set_uint32),
3869 (dbus_dict_set_double), (dbus_dict_set_string),
3870 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
3871 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
3872 (dbus_dict_set_string_array), (_dbus_dict_test):
3874 Fix according to comments from Havoc.
3876 2003-03-06 Michael Meeks <michael@server.home>
3878 * configure.in: if we don't have kde-config, disable have_qt.
3880 2003-03-07 Anders Carlsson <andersca@codefactory.se>
3885 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
3886 (dbus_dict_ref), (dbus_dict_unref), (dbus_dict_contains),
3887 (dbus_dict_remove), (dbus_dict_get_value_type),
3888 (dbus_dict_get_keys), (dbus_dict_put_boolean),
3889 (dbus_dict_put_int32), (dbus_dict_put_uint32),
3890 (dbus_dict_put_double), (dbus_dict_put_string),
3891 (dbus_dict_put_boolean_array), (dbus_dict_put_int32_array),
3892 (dbus_dict_put_uint32_array), (dbus_dict_put_double_array),
3893 (dbus_dict_put_string_array), (dbus_dict_get_boolean),
3894 (dbus_dict_get_int32), (dbus_dict_get_uint32),
3895 (dbus_dict_get_double), (dbus_dict_get_string),
3896 (dbus_dict_get_boolean_array), (dbus_dict_get_int32_array),
3897 (dbus_dict_get_uint32_array), (dbus_dict_get_double_array),
3898 (dbus_dict_get_string_array), (_dbus_dict_test):
3900 Add DBusDict implementation
3902 * dbus/dbus-test.c: (dbus_internal_do_not_use_run_tests):
3906 2003-03-04 Havoc Pennington <hp@pobox.com>
3908 * test/data/auth/*: adapt to changes
3910 * dbus/dbus-auth-script.c (_dbus_auth_script_run): add
3911 USERID_BASE64 and change USERNAME_BASE64 to put in username not
3914 * dbus/dbus-keyring.c (_dbus_keyring_validate_context): prevent
3915 more stuff from being in a context name, to make the protocol
3916 simpler to deal with
3918 * dbus/dbus-errors.c (dbus_error_has_name): new function
3919 (dbus_error_is_set): new function
3921 * dbus/dbus-auth.c: replace DBUS_STUPID_TEST_MECH auth
3922 with DBUS_COOKIE_SHA1, implement DBUS_COOKIE_SHA1
3924 * dbus/dbus-connection.c (dbus_connection_flush): also read
3925 messages during a flush operation
3927 * dbus/Makefile.am: remove dbus-md5 since it isn't currently used.
3929 2003-03-05 Anders Carlsson <andersca@codefactory.se>
3931 * configure.in: Check for gethostbyname on Solaris.
3933 * dbus/dbus-transport.c: (_dbus_transport_open):
3934 Remove duplicate "tcp" entry.
3936 * doc/dbus-specification.sgml:
3937 Clarify some things.
3939 2003-03-05 Anders Carlsson <andersca@codefactory.se>
3941 * dbus/dbus-auth.c: (send_rejected), (process_test_subdir):
3942 * dbus/dbus-keyring.c: (_dbus_keyring_new_homedir),
3943 (_dbus_keyring_test):
3944 * dbus/dbus-md5.c: (_dbus_md5_compute):
3945 * dbus/dbus-sha.c: (_dbus_sha_compute):
3948 2003-03-05 Anders Carlsson <andersca@codefactory.se>
3950 * README: Add some things.
3952 2003-03-04 Anders Carlsson <andersca@codefactory.se>
3954 * dbus/dbus-message.c (dbus_message_append_args_valist): Add a break;
3955 after case DBUS_TYPE_BOOELAN.
3957 2003-03-02 Havoc Pennington <hp@pobox.com>
3959 * test/break-loader.c (randomly_set_extreme_ints): add test that
3960 sets really huge and small integers
3962 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add check
3963 that length of boolean array fits in the string, and that
3964 string has room for boolean value in single-bool case.
3966 * dbus/dbus-message-builder.c (_dbus_message_data_load): add
3967 optional value to "ALIGN" command which is what to fill the
3970 * test/data/valid-messages/no-padding.message: add regression
3971 test for the message padding problem
3973 2003-03-02 Havoc Pennington <hp@pobox.com>
3975 * dbus/dbus-message.c (decode_header_data): fix to always init
3976 message_padding, from Benjamin Dauvergne
3978 2003-03-02 Havoc Pennington <hp@pobox.com>
3984 2003-03-01 Joe Shaw <joe@assbarn.com>
3986 * configure.in: Check for "struct cmsgcred" and try to access its
3987 members for BSD-like unices.
3989 * dbus/dbus-sysdeps.c (read_credentials_byte): Fold this back into
3990 _dbus_read_credentials_unix_socket().
3991 (_dbus_read_credentials_unix_socket): Use recvmsg() instead of
3992 read() for reading the credential byte off the unix socket. Use
3993 struct cmsgcred on systems that support it.
3995 2003-02-27 Alexander Larsson <alexl@redhat.com>
3999 Make gthreads-2.0 dependency optional. Don't build thread test if
4002 2003-02-27 Havoc Pennington <hp@pobox.com>
4004 * dbus/dbus-connection.c
4005 (dbus_connection_send_message_with_reply_and_block): fix doh!
4006 doh! doh! bug that resulted in never removing a reply from the
4007 queue, no wonder we called get_reply_serial so much ;-)
4009 * dbus/dbus-message.c (struct DBusMessage): cache reply serial
4010 and client serial instead of demarshaling them every time
4012 2003-02-27 Havoc Pennington <hp@pobox.com>
4014 * dbus/dbus-marshal.c (_dbus_demarshal_int32): rewrite to be much
4015 more inlined, using dbus-string-private.h, speeds things up
4018 * dbus/dbus-string.c (_dbus_string_free): apply align offset
4019 when freeing the string
4020 (_dbus_string_steal_data): fix for align offset
4021 (undo_alignment): new function
4023 2003-02-26 Havoc Pennington <hp@redhat.com>
4025 All kinds of audit fixes from Owen, plus initial attempt to
4026 handle unaligned memory returned from malloc.
4028 * dbus/dbus-string.c (_dbus_string_init): clamp max length to
4029 leave room for align_offset and nul byte
4030 (fixup_alignment): function to track an align_offset and
4031 ensure real->str is aligned
4032 (DBUS_GENERIC_STRING_PREAMBLE): len must be less than allocated,
4033 to allow a nul byte plus align offset
4034 (_dbus_string_lock): fix overflow issue
4035 (_dbus_string_init_const_len): add assertions on sanity of len,
4036 assign allocated to be ALLOCATION_PADDING larger than len
4037 (set_length): fixup the overflow handling
4038 (_dbus_string_get_data_len): fix overflow in assertion
4039 (open_gap): detect overflow in size of gap to be opened
4040 (_dbus_string_lengthen): add overflow check
4041 (_dbus_string_align_length): fix overflow with _DBUS_ALIGN_VALUE
4042 (_dbus_string_append): add overflow check
4043 (_dbus_string_append_unichar): overflow
4044 (_dbus_string_delete): fix overflow in assertion
4045 (_dbus_string_copy_len): overflow in assertion
4046 (_dbus_string_replace_len): overflows in assertions
4047 (_dbus_string_find): change to implement in terms of
4048 _dbus_string_find_to
4049 (_dbus_string_find_to): assorted fixage
4050 (_dbus_string_equal_c_str): assert c_str != NULL,
4051 fix logic so the function works
4052 (_dbus_string_ends_with_c_str): fix overflow thingy
4053 (_dbus_string_base64_encode): overflow fix
4054 (_dbus_string_validate_ascii): overflow
4055 (_dbus_string_validate_nul): overflow
4057 2003-02-26 Havoc Pennington <hp@redhat.com>
4059 * dbus/dbus-marshal.c (_dbus_marshal_test): fix to work with DISABLE_ASSERTS
4061 2003-02-26 Alexander Larsson <alexl@redhat.com>
4064 Set DBUS_GLIB_THREADS_LIBS for apps using gthread-2.0
4066 * dbus/dbus-connection.c:
4067 * dbus/dbus-connection.h:
4068 Fix _dbus_connection_acquire_io_path and _dbus_connection_acquire_dispatch.
4069 Add dbus_connection_set_wakeup_main_function and use it when queueing
4070 incoming and outgoing messages.
4073 * dbus/dbus-dataslot.c:
4074 Threadsafe usage of DBusDataSlotAllocator
4076 * dbus/dbus-message.c: (dbus_message_get_args_iter):
4079 * dbus/dbus-server-unix.c:
4082 * glib/dbus-gmain.c:
4083 Implement the new wakeup functions for glib.
4086 * glib/test-thread-client.c:
4087 * glib/test-thread-server.c:
4088 * glib/test-thread.h:
4089 Initial cut at some thread test code. Not really done yet.
4091 2003-02-26 Havoc Pennington <hp@pobox.com>
4093 * dbus/dbus-connection.c
4094 (dbus_connection_send_message_with_reply_and_block): fix crash
4095 where we ref'd the outgoing message instead of the returned reply
4097 * dbus/dbus-transport-unix.c (do_authentication): check read watch
4098 at the end of this function, so if we didn't need to read for
4099 authentication, we reinstall it for receiving messages
4101 * dbus/dbus-message.c (dbus_message_new_reply): allow replies to
4102 a NULL sender for peer-to-peer case
4104 * dbus/dbus-transport-unix.c (check_read_watch): handle
4105 !authenticated case correctly
4107 * glib/dbus-gmain.c: add support for DBusServer
4109 * dbus/dbus-server.c: add data slot support
4111 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): check
4112 return values and handle errors
4114 * dbus/dbus-dataslot.c: factor out the data slot stuff from
4117 * Doxyfile.in (INPUT): add glib subdir
4119 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): rename
4120 setup_with_g_main instead of hookup_with_g_main; write docs
4122 2003-02-24 Anders Carlsson <andersca@codefactory.se>
4124 * dbus/dbus-marshal.c: (_dbus_marshal_validate_arg):
4125 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
4126 * dbus/dbus-message.c: (dbus_message_append_boolean),
4127 (dbus_message_append_boolean_array),
4128 (dbus_message_get_args_valist), (_dbus_message_test):
4129 * dbus/dbus-message.h:
4130 * doc/dbus-specification.sgml:
4131 Various fixes as pointed out by Havoc.
4133 * test/data/invalid-messages/bad-boolean-array.message:
4134 * test/data/invalid-messages/bad-boolean.message:
4135 Add invalid boolean value test cases.
4137 2003-02-24 Anders Carlsson <andersca@codefactory.se>
4139 * dbus/dbus-internals.c: (_dbus_type_to_string):
4140 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
4141 (_dbus_marshal_validate_arg):
4142 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
4143 * dbus/dbus-message.c: (dbus_message_append_args_valist),
4144 (dbus_message_append_boolean), (dbus_message_append_boolean_array),
4145 (dbus_message_get_args_valist), (dbus_message_iter_get_boolean),
4146 (dbus_message_iter_get_int32), (dbus_message_iter_get_uint32),
4147 (dbus_message_iter_get_double),
4148 (dbus_message_iter_get_boolean_array), (message_iter_test):
4149 * dbus/dbus-message.h:
4150 * dbus/dbus-protocol.h:
4151 * doc/dbus-specification.sgml:
4152 * test/data/valid-messages/lots-of-arguments.message:
4153 Add support for boolean and boolean array types.
4155 2003-02-23 Havoc Pennington <hp@pobox.com>
4157 * dbus/dbus-keyring.c: finish most of this implementation and
4160 * dbus/dbus-errors.c (dbus_set_error_const, dbus_set_error): make
4161 these barf if the error isn't cleared to NULL
4163 * dbus/dbus-sysdeps.c (_dbus_delete_file): set error on failure
4164 (_dbus_create_directory): new function
4166 * dbus/dbus-errors.c (dbus_set_error): fix warning
4168 * dbus/dbus-string.c (_dbus_string_hex_encode): new function
4169 (_dbus_string_hex_decode): new function
4170 (test_hex_roundtrip): test code
4172 * dbus/dbus-sha.c (_dbus_sha_compute): use dbus_string_hex_encode
4174 * dbus/dbus-md5.c (_dbus_md5_compute): use dbus_string_hex_encode
4176 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): make this use
4177 the save-to-temp/rename trick to atomically write the new file
4178 (_dbus_string_parse_uint): new function
4180 2003-02-22 Havoc Pennington <hp@pobox.com>
4182 * test/Makefile.am (dist-hook): fix dist for test/data/sha-1
4184 2003-02-22 Havoc Pennington <hp@pobox.com>
4186 * dbus/dbus-message.c (dbus_message_iter_get_string_array):
4187 (dbus_message_iter_get_byte_array): Fix up doxygen warnings
4189 * dbus/dbus-sha.c: add implementation of SHA-1 algorithm
4191 * dbus/test/data/sha-1: add US government test suite for SHA-1
4193 2003-02-21 Anders Carlsson <andersca@codefactory.se>
4195 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
4196 Make string arrays NULL-terminated.
4198 * dbus/dbus-memory.c: (dbus_free_string_array):
4199 * dbus/dbus-memory.h:
4200 New function for freeing NULL-terminated string arrays.
4202 * dbus/dbus-message-builder.c: (append_quoted_string),
4203 (_dbus_message_data_load):
4204 Add support for array types.
4206 * dbus/dbus-message.c: (check_message_handling):
4207 Add more types as test cases.
4209 * dbus/dbus-sysdeps.c: (_dbus_string_parse_int),
4210 (_dbus_string_parse_double):
4211 Add the start offset to the end offset.
4213 * test/data/valid-messages/lots-of-arguments.message:
4214 New test message with lots of arguments.
4216 2003-02-21 Anders Carlsson <andersca@codefactory.se>
4218 * dbus/dbus-message.c: (dbus_message_append_nil),
4219 (dbus_message_append_int32), (dbus_message_append_uint32),
4220 (dbus_message_append_double), (dbus_message_append_string),
4221 (dbus_message_append_int32_array),
4222 (dbus_message_append_uint32_array),
4223 (dbus_message_append_double_array),
4224 (dbus_message_append_byte_array),
4225 (dbus_message_append_string_array):
4226 Fix all out-of-memory handling in these functions.
4228 2003-02-21 Anders Carlsson <andersca@codefactory.se>
4230 * dbus/dbus-message.c: (dbus_message_append_nil):
4233 2003-02-21 Anders Carlsson <andersca@codefactory.se>
4235 * dbus/dbus-message.c: (dbus_message_append_args_valist),
4236 (dbus_message_append_nil), (dbus_message_append_int32_array),
4237 (dbus_message_append_uint32_array),
4238 (dbus_message_append_double_array),
4239 (dbus_message_append_byte_array),
4240 (dbus_message_append_string_array), (dbus_message_get_args_valist),
4241 (dbus_message_iter_get_int32_array),
4242 (dbus_message_iter_get_uint32_array),
4243 (dbus_message_iter_get_double_array),
4244 (dbus_message_iter_get_byte_array),
4245 (dbus_message_iter_get_string_array):
4247 * dbus/dbus-message.h:
4248 Add functions for appending and getting arrays.
4250 2003-02-21 Anders Carlsson <andersca@codefactory.se>
4252 * dbus/dbus-mempool.c (_dbus_mem_pool_new): Make the
4253 element size at least 8 bytes, fixes mempool tests on
4256 2003-02-20 Alexander Larsson <alexl@redhat.com>
4258 * dbus/dbus-transport-unix.c (unix_do_iteration):
4259 Unlock the connection mutex during a blocking select call.
4260 Add todo about how we need a way to wake up the select.
4262 * dbus/dbus-connection-internal.h:
4263 * dbus/dbus-connection.c:
4264 Add _dbus_connection_lock and _dbus_connection_unlock.
4266 2003-02-19 Havoc Pennington <hp@pobox.com>
4268 * Doxyfile.in (PREDEFINED): put DOXYGEN_SHOULD_SKIP_THIS in
4269 Doxyfile.in, not Doxyfile
4271 * dbus/dbus-keyring.c: do some hacking on this
4273 * dbus/dbus-sysdeps.c (_dbus_delete_file): new
4275 * dbus/dbus-errors.c (dbus_set_error_const): do not call
4277 (dbus_set_error): remove dbus_error_init, check for message ==
4278 NULL *before* we sprintf into it, and add @todo about including
4279 system headers in this file
4281 * dbus/dbus-sysdeps.c (_dbus_create_file_exclusively): new
4283 * dbus/dbus-errors.h (DBUS_ERROR_FAILED): add
4285 * dbus/dbus-sysdeps.c (get_user_info): break this function out to
4286 get various bits of user information based on either username
4288 (_dbus_homedir_from_username): new function
4290 2003-02-19 Anders Carlsson <andersca@codefactory.se>
4293 Add check for nonposix getpwnam_r
4295 * dbus/dbus-mempool.c: (_dbus_mem_pool_new):
4296 Align the pool element size to a sizeof (void *) boundary.
4298 * dbus/dbus-sysdeps.c: (_dbus_setenv), (_dbus_connect_unix_socket),
4299 (_dbus_listen_unix_socket), (_dbus_credentials_from_username):
4300 General Solaris fixes.
4302 * test/data/valid-messages/simplest-manual.message:
4303 Explicitly state that we want little-endian packing.
4305 2003-02-19 Mikael Hallendal <micke@codefactory.se>
4307 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
4309 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_tcp_socket):
4310 Added to create a transport connecting using a tcp/ip socket.
4312 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): Added to connect
4313 to a tcp socket at given host and port.
4314 (_dbus_listen_tcp_socket): added to listen on tcp socket for given
4317 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
4319 * dbus/dbus-server-unix.c (_dbus_server_new_for_tcp_socket):
4320 Added to create a server listening on a TCP/IP socket.
4322 2003-02-19 Havoc Pennington <hp@pobox.com>
4324 Throughout: mop up all the Doxygen warnings and undocumented
4327 * dbus/dbus-sysdeps.c (do_exec): do not use execvp, we don't want
4328 to search any paths.
4330 * dbus/dbus-threads.c: move global mutex initializers to
4331 dbus-internals.h, multiple prototypes was confusing doxygen
4332 besides being kind of ugly
4334 * Doxyfile (PREDEFINED): have Doxygen define
4335 DOXYGEN_SHOULD_SKIP_THIS so we can exclude things from
4336 docs with #ifndef DOXYGEN_SHOULD_SKIP_THIS
4337 (do not abuse the feature! it's for stuff like the autogenerated
4338 macros in dbus-md5.c, not just for things you don't feel like
4341 2003-02-18 Havoc Pennington <hp@pobox.com>
4343 * dbus/dbus-string.c (_dbus_string_zero): new function
4345 * dbus/dbus-md5.c: include MD5 implementation by L. Peter Deutsch,
4346 wrap it in some dbus-friendly API
4348 * dbus/dbus-types.h: add 16-bit types
4350 2003-02-18 Joe Shaw <joe@assbarn.com>
4352 * dbus/dbus-auth.c (handle_server_data_stupid_test_mech): Just get
4353 credentials from our currently running process.
4354 (get_word): Fix a buglet where we were copying the entire length
4355 instead of relative to our position.
4357 * dbus/dbus-hash.c (_dbus_hash_test): Don't try to allocate the
4358 keys on the stack... it's 640k of data.
4360 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): Always
4361 read the credentials byte off the socket, even if we don't have
4363 (_dbus_poll): Implement poll() using select() for systems which
4366 * glib/test-dbus-glib.c (main): Print out an error if no
4367 parameters are given.
4369 * test/data/auth/fallback.auth-script: Added. Tests that a client
4370 can fallback to a secondary auth mechanism if the first fails.
4372 2003-02-18 Havoc Pennington <hp@pobox.com>
4376 2003-02-17 Havoc Pennington <hp@pobox.com>
4378 * doc/dbus-specification.sgml: lots of cosmetic
4379 cleanups/rearrangement, add assorted FIXME, change DBUS_ADDRESS
4380 env variable to DBUS_BUS_ADDRESS, s/client/application/,
4381 s/server/bus/ (except in authentication section). Add a section
4382 "Message Bus Message Routing"
4384 2003-02-17 Anders Carlsson <andersca@codefactory.se.>
4390 2003-02-17 Anders Carlsson <andersca@codefactory.se>
4392 * doc/dbus-specification.sgml:
4393 Specification updates.
4395 2003-02-17 Anders Carlsson <andersca@codefactory.se>
4397 * bus/activation.c: (bus_activation_init), (child_setup),
4398 (bus_activation_activate_service):
4400 * bus/main.c: (main):
4401 Set DBUS_ADDRESS environment variable.
4403 * dbus/dbus-errors.c: (dbus_set_error):
4404 Don't use va_copy since that's a C99 feature.
4406 * dbus/dbus-sysdeps.c: (_dbus_setenv), (do_exec),
4407 (_dbus_spawn_async):
4408 * dbus/dbus-sysdeps.h:
4409 Add child_setup_func to _dbus_spawn_async.
4411 * doc/dbus-specification.sgml:
4412 Update specification.
4414 * test/spawn-test.c: (setup_func), (main):
4417 2003-02-17 Alexander Larsson <alexl@redhat.com>
4419 * dbus/dbus-connection.c (_dbus_connection_handler_destroyed_locked):
4422 2003-02-17 Anders Carlsson <andersca@codefactory.se>
4426 * doc/dbus-test-plan.sgml:
4427 Add test plan document.
4432 2003-02-17 Anders Carlsson <andersca@codefactory.se>
4434 * dbus/dbus-message.c: (decode_header_data),
4435 (_dbus_message_loader_return_buffer):
4436 Set the header padding amount when loading a message.
4438 2003-02-16 Anders Carlsson <andersca@codefactory.se>
4440 * bus/dispatch.c: (send_one_message):
4441 Only send broadcast messages to registered connections.
4443 * dbus/dbus-message.c: (dbus_message_name_is):
4444 * dbus/dbus-message.h:
4445 New convenience function.
4447 * dbus/dbus-transport-debug.c: (do_reading):
4448 Only dispatch one message per run.
4451 * test/bus-test.c: (new_connection_callback), (die),
4452 (test_hello_client1_handler), (test_hello_client2_handler),
4453 (test_hello_replies), (main):
4455 * test/bus-test-loop.[ch]:
4458 2003-02-16 Havoc Pennington <hp@pobox.com>
4460 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
4461 backward conditional
4463 2003-02-16 Alexander Larsson <alexl@redhat.com>
4465 * dbus/dbus-connection.c:
4466 Implement sent_message_with_reply. (with_reply_and block is still
4468 Made dispatch_message not lose message if OOM.
4470 * dbus/dbus-errors.h:
4471 Add NoReply error (for reply timeouts).
4473 2003-02-16 Alexander Larsson <alexl@redhat.com>
4475 * dbus/dbus-hash.c (_dbus_hash_table_unref):
4476 Actually free keys and values when destroying hashtable.
4478 2003-02-16 Anders Carlsson <andersca@codefactory.se>
4480 * dbus/dbus-auth.c: (client_try_next_mechanism):
4483 * dbus/dbus-threads.c: (dbus_condvar_wait_timeout):
4484 Return TRUE if there's no thread implementation around.
4486 * glib/dbus-gmain.c: (free_source),
4487 (dbus_connection_hookup_with_g_main):
4488 Make sure to remove the GSource when the connection is finalized.
4490 2003-02-16 Anders Carlsson <andersca@codefactory.se>
4492 * bus/dispatch.c: (bus_dispatch_message_handler):
4493 * dbus/dbus-errors.h:
4494 Return an error if someone tries to send a message to a service
4497 2003-02-16 Anders Carlsson <andersca@codefactory.se>
4499 * bus/activation.c: (load_directory), (bus_activation_init),
4500 (bus_activation_activate_service):
4503 (bus_driver_handle_activate_service), (bus_driver_handle_message):
4504 More work on the activation handling.
4506 * dbus/dbus-errors.h:
4507 Add some error messages
4509 * dbus/dbus-message.c: (dbus_message_new_error_reply):
4510 * dbus/dbus-message.h:
4511 New function that creates an error message.
4513 * dbus/dbus-protocol.h:
4514 Add ACTIVATE_SERVER message.
4516 * dbus/dbus-server-unix.c: (unix_handle_watch),
4517 (_dbus_server_new_for_domain_socket):
4518 Call _dbus_fd_set_close_on_exec.
4520 * dbus/dbus-sysdeps.c: (make_pipe), (do_exec),
4521 (_dbus_spawn_async), (_dbus_disable_sigpipe),
4522 (_dbus_fd_set_close_on_exec):
4523 * dbus/dbus-sysdeps.h:
4524 Add _dbus_fd_set_close_on exec function. Also add function that checks
4525 that all open fds are set to close-on-exec and warns otherwise.
4527 * dbus/dbus-transport-unix.c:
4528 (_dbus_transport_new_for_domain_socket):
4529 Call _dbus_fd_set_close_on_exec.
4531 2003-02-16 Havoc Pennington <hp@pobox.com>
4533 * dbus/dbus-connection.c (dbus_connection_set_change_sigpipe):
4534 allow people to avoid setting SIGPIPE to SIG_IGN
4535 (_dbus_connection_new_for_transport): disable SIGPIPE unless
4536 we've been asked not to
4538 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4540 * dbus/dbus-list.c: (_dbus_list_append_link),
4541 (_dbus_list_prepend_link):
4542 * dbus/dbus-memory.c: (dbus_malloc), (dbus_malloc0),
4546 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4549 * bus/activation.c: (bus_activation_entry_free),
4550 (add_desktop_file_entry), (load_directory), (bus_activation_init):
4552 * bus/main.c: (main):
4553 Add simple activation support, doesn't work yet though.
4555 2003-02-15 Zack Rusin <zack@kde.org>
4557 * qt/dbus-qthread.cpp: small casting fix
4559 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4561 * dbus/dbus-errors.c: (dbus_set_error):
4562 * dbus/dbus-errors.h:
4563 Add a few errors and make dbus_set_error void.
4565 * dbus/dbus-sysdeps.c:
4566 (_dbus_errno_to_string), (close_and_invalidate), (make_pipe),
4567 (write_err_and_exit), (read_ints), (do_exec), (_dbus_spawn_async):
4568 * dbus/dbus-sysdeps.h:
4569 Add _dbus_spawn_async.
4571 * test/spawn-test.c: (main):
4572 Test for _dbus_spawn_async.
4574 2003-02-15 Anders Carlsson <andersca@codefactory.se>
4576 * dbus/dbus-internals.h:
4577 Fix build without tests.
4579 * dbus/dbus-list.c: (alloc_link):
4580 Fix a segfault when a malloc fails.
4582 * dbus/dbus-memory.c: (initialize_malloc_debug), (dbus_malloc),
4583 (dbus_malloc0), (dbus_realloc):
4584 Add support for malloc debugging.
4586 2003-02-15 Alexander Larsson <alexl@redhat.com>
4588 * dbus/dbus-threads.c:
4589 * dbus/dbus-threads.h:
4590 Add condvars. Remove static mutext from API.
4591 Implement static mutexes by initializing them from threads_init.
4593 * glib/dbus-gthread.c:
4594 * qt/dbus-qthread.cpp:
4595 Update with the thread api changes.
4600 Turn StaticMutex into normal mutex + init function.
4601 Export new functions _dbus_list_alloc_link, _dbus_list_free_link,
4602 _dbus_list_append_link, _dbus_list_prepend_link
4605 * dbus/dbus-sysdeps.c:
4606 * dbus/dbus-sysdeps.h:
4607 New type dbus_atomic_t, and new functions _dbus_atomic_inc,
4608 _dbus_atomic_dec. Only slow fallback implementation at the moment.
4610 * dbus/dbus-protocol.h:
4611 Add DBUS_MESSAGE_LOCAL_DISCONNECT define
4613 * dbus/dbus-message.c:
4614 Make ref/unref atomic.
4617 * dbus/dbus-connection-internal.h:
4618 * dbus/dbus-connection.c:
4619 * dbus/dbus-connection.h:
4621 Change _peek to _borrow,_return & _steal_borrowed.
4622 Change disconnect callback to event.
4623 Make dbus_connection_dispatch_messages reentrant.
4625 * dbus/dbus-transport.c:
4626 Don't ref the connection on calls to the transport
4629 * dbus/dbus-message-handler.c:
4632 * glib/dbus-gmain.c:
4633 Don't use peek_message anymore
4636 * test/debug-thread.c:
4637 * test/debug-thread.h:
4638 Simple thread implementation that asserts() on deadlocks in
4639 single-threaded code.
4642 (main) Call debug_threads_init.
4645 Use disconnect message instead of disconnect callback.
4649 Don't call dbus_connection_set_disconnect_function. Instead export
4650 bus_connection_disconnect.
4653 Call bus_connection_disconnect when we get a disconnected message.
4655 2003-02-15 Havoc Pennington <hp@pobox.com>
4657 * dbus/dbus-message.c (dbus_message_new): fool around with the
4660 2003-02-14 Havoc Pennington <hp@pobox.com>
4662 * dbus/dbus-mempool.c: fail if the debug functions so indicate
4664 * dbus/dbus-memory.c: fail if the debug functions indicate we
4667 * dbus/dbus-internals.c (_dbus_set_fail_alloc_counter)
4668 (_dbus_decrement_fail_alloc_counter): debug functions to
4669 simulate memory allocation failures
4671 2003-02-14 Havoc Pennington <hp@pobox.com>
4673 * dbus/dbus-errors.h (struct DBusError): add a word of padding
4676 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4678 * bus/driver.c: (bus_driver_handle_hello):
4680 * bus/services.c: (bus_service_lookup):
4681 Reorder message sending so we get a more sane order.
4683 * test/bus-test.c: (message_handler):
4686 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4688 * bus/driver.c: (bus_driver_send_service_deleted),
4689 (bus_driver_send_service_created), (bus_driver_send_service_lost),
4690 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
4691 (bus_driver_send_welcome_message),
4692 (bus_driver_handle_list_services),
4693 (bus_driver_handle_acquire_service),
4694 (bus_driver_handle_service_exists):
4695 * dbus/dbus-bus.c: (dbus_bus_register_client),
4696 (dbus_bus_acquire_service), (dbus_bus_service_exists):
4697 * dbus/dbus-errors.c: (dbus_result_to_string):
4698 * dbus/dbus-errors.h:
4699 * dbus/dbus-message.c: (dbus_message_append_args),
4700 (dbus_message_append_args_valist), (dbus_message_get_args),
4701 (dbus_message_get_args_valist), (dbus_message_get_args_iter),
4702 (dbus_message_iter_get_arg_type), (dbus_message_iter_get_string),
4703 (dbus_message_iter_get_byte_array),
4704 (dbus_message_iter_get_string_array), (message_iter_test),
4705 (check_message_handling), (_dbus_message_test):
4706 * dbus/dbus-message.h:
4707 * test/bus-test.c: (main):
4708 Change fields to arguments in messages, so that they won't be
4709 confused with header fields.
4711 * glib/test-dbus-glib.c: (main):
4712 Remove append_fields from hello message.
4714 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4716 * dbus/dbus-errors.c:
4717 * dbus/dbus-message.c:
4718 * dbus/dbus-string.c:
4719 Documentation fixes.
4721 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4723 * glib/dbus-gmain.c: (timeout_handler), (add_timeout),
4725 Implement support for timeouts in dbus-glib.
4727 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4729 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
4730 * dbus/dbus-message.c: (process_test_subdir):
4731 * test/break-loader.c: (find_breaks_based_on):
4732 Plug some memory leaks.
4734 2003-02-13 Richard Hult <rhult@codefactory.se>
4736 * bus/main.c: Fix build.
4738 * dbus/dbus-errors.h:
4739 * dbus/dbus-errors.c: Fix copyright for Anders.
4741 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4746 * bus/connection.c: (bus_connection_foreach):
4749 * bus/desktop-file.c: (grow_lines_in_section), (grow_sections),
4750 (unescape_string), (new_section), (parse_section_start),
4751 (parse_key_value), (report_error), (bus_desktop_file_load),
4752 (bus_desktop_file_get_string):
4753 * bus/desktop-file.h:
4754 Use DBusError for error reporting.
4756 * bus/dispatch.c: (send_one_message),
4757 (bus_dispatch_message_handler):
4758 * bus/driver.c: (bus_driver_send_service_deleted),
4759 (bus_driver_send_service_created), (bus_driver_send_service_lost),
4760 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
4761 (bus_driver_send_welcome_message),
4762 (bus_driver_handle_list_services),
4763 (bus_driver_handle_acquire_service),
4764 (bus_driver_handle_service_exists):
4765 * bus/loop.c: (bus_loop_run):
4767 Use BUS_HANDLE_OOM instead of _DBUS_HANDLE_OOM.
4769 * bus/utils.c: (bus_wait_for_memory):
4771 New files with general utility functions.
4773 * dbus/dbus-internals.h:
4774 Remove _DBUS_HANDLE_OOM.
4776 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4778 * dbus/dbus-errors.c: (dbus_result_to_string), (dbus_error_init),
4779 (dbus_error_free), (dbus_set_error_const), (dbus_set_error):
4780 * dbus/dbus-errors.h:
4781 Add DBusError structure.
4783 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4785 * test/data/valid-messages/standard-acquire-service.message:
4786 * test/data/valid-messages/standard-hello.message:
4787 * test/data/valid-messages/standard-list-services.message:
4788 * test/data/valid-messages/standard-service-exists.message:
4789 Add some standard messages.
4791 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4793 * bus/driver.c: (bus_driver_send_welcome_message),
4794 (bus_driver_handle_list_services),
4795 (bus_driver_handle_acquire_service),
4796 (bus_driver_handle_service_exists), (bus_driver_handle_message):
4797 Update for API changes in libdbus.
4799 * dbus/dbus-message.c: (dbus_message_new_reply):
4800 * dbus/dbus-message.h:
4801 Remove the name argument. The spec states that replies shouldn't
4804 2003-02-13 Anders Carlsson <andersca@codefactory.se>
4806 * bus/desktop-file.c: (parse_section_start), (parse_key_value),
4807 (report_error), (bus_desktop_file_load), (lookup_section),
4808 (lookup_line), (bus_desktop_file_get_raw),
4809 (bus_desktop_file_get_string):
4810 * bus/desktop-file.h:
4811 Some fixes, and new functions for getting a key value from a section.
4813 2003-02-13 Havoc Pennington <hp@pobox.com>
4815 * test/data/auth/fail-after-n-attempts.auth-script: new test
4817 * dbus/dbus-auth.c (send_rejected): shutdown_mech() when we
4820 2003-02-13 Havoc Pennington <hp@pobox.com>
4822 * dbus/dbus-auth.c (handle_server_data_external_mech): args to
4823 dbus_credentials_match were backward
4825 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support
4826 NO_CREDENTIALS and ROOT_CREDENTIALS
4828 * dbus/dbus-auth.c (_dbus_auth_do_work): move get_state() routine
4829 into here. Never process more commands after we've reached an
4830 end state; store further data as unused bytes.
4832 * test/data/auth/*: add more auth tests
4834 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support EXPECT
4835 command to match exact string and EXPECT_UNUSED to match unused
4838 * test/Makefile.am (dist-hook): fix to dist all the test stuff
4840 2003-02-12 Havoc Pennington <hp@pobox.com>
4842 * dbus/dbus-string.c (_dbus_string_pop_line): fix to also strip
4843 \r off of popped lines
4845 * dbus/dbus-auth.c (_dbus_auth_test): write code to run auth
4848 * dbus/dbus-auth-script.c (_dbus_auth_script_run): when doing a
4851 2003-02-12 Havoc Pennington <hp@pobox.com>
4853 * dbus/Makefile.am: remove break-loader from the build, since it
4856 * configure.in: add --enable-gcov to turn on coverage profiling
4857 flags and disable optimization
4859 2003-02-10 Havoc Pennington <hp@pobox.com>
4861 * dbus/dbus-auth-script.c, dbus/dbus-auth-script.h: sync
4862 initial cut at test framework for DBusAuth from laptop.
4863 Doesn't quite work yet but it compiles and I need to get
4864 it off the 266mhz laptop. ;-)
4866 * dbus/dbus-server-debug.c (_dbus_server_debug_accept_transport):
4867 fix a memleak in error case
4869 2003-02-12 Anders Carlsson <andersca@codefactory.se>
4872 * bus/desktop-file.c:
4873 * bus/desktop-file.h:
4874 Add a desktop file parser.
4876 2003-02-11 Zack Rusin <zack@kde.org>
4878 * qt/message.[h|cpp]: sample implementation
4879 of the KDE wrapper for DBusMessage
4881 2003-02-09 Zack Rusin <zack@kde.org>
4883 * test/bus-test.c: make_it_compile
4884 * doc/dbus-specification.sgml: minimal semantic fix
4886 2003-02-06 Anders Carlsson <andersca@codefactory.se>
4892 2003-02-06 Anders Carlsson <andersca@codefactory.se>
4895 * dbus/dbus-break-loader.c:
4897 * test/break-loader.c:
4898 Move dbus-break-loader to test/ and rename it to break-loader.
4900 2003-02-02 Havoc Pennington <hp@pobox.com>
4902 * dbus/dbus-keyring.c, dbus/dbus-keyring.h: template files
4903 for code to manage cookies in your home directory
4905 * dbus/dbus-sysdeps.c (_dbus_generate_random_bytes): new function
4907 * dbus/dbus-auth.c (get_state): impose a maximum number of tries
4908 to authenticate, then disconnect the client.
4910 2003-02-03 Alexander Larsson <alexl@redhat.com>
4912 * dbus/dbus-message.c (dbus_message_append_fields):
4915 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4917 * doc/dbus-specification.sgml:
4918 Update address format section.
4920 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4923 * test/bus-test.c: (get_time), (add_timeout), (remove_timeout),
4924 (message_handler), (new_connection_callback), (loop_quit),
4928 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4930 * bus/driver.c: (bus_driver_handle_service_exists):
4931 Simplify the code a bit.
4933 * dbus/dbus-bus.c: (dbus_bus_service_exists):
4936 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4939 Add libdbus-daemon.la and link to it.
4941 2003-02-01 James Willcox <jwillcox@gnome.org>
4943 * bus/driver.c: (bus_driver_handle_own_service):
4944 Actually include the service reply code in the message.
4946 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4948 * bus/driver.c: (bus_driver_handle_service_exists):
4949 Don't unref the incoming message.
4951 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4953 * dbus/dbus.h: Add dbus-address.h and dbus-bus.h
4955 2003-02-02 Anders Carlsson <andersca@codefactory.se>
4957 * dbus/dbus-server.c: (dbus_server_listen):
4958 * dbus/dbus-transport.c: (_dbus_transport_open):
4959 ifdef out the calls to the debug transport and server.
4961 2003-02-02 Alexander Larsson <alexl@redhat.com>
4963 * dbus/dbus-watch.c (dbus_watch_get_flags):
4964 Add note in the docs that ERROR or HANGUP won't be returned
4965 and are assumed always on.
4967 * glib/dbus-gmain.c (add_watch):
4968 Always add IO_ERR | IO_HUP
4970 * dbus/dbus-message.h:
4971 Add semicolon after dbus_message_iter_get_string_array().
4972 Makes qt code build again
4974 2003-02-01 Anders Carlsson <andersca@codefactory.se>
4976 * bus/driver.c: (create_unique_client_name),
4977 (bus_driver_handle_hello):
4978 Don't take a name, just use a numeric id to identify
4982 * dbus/dbus-bus.c: (dbus_bus_register_client),
4983 (dbus_bus_acquire_service), (dbus_bus_service_exists):
4985 Add new convenience functions for communicating with the bus.
4987 * dbus/dbus-message.h:
4989 * dbus/dbus-protocol.h:
4992 2003-02-01 Alexander Larsson <alexl@redhat.com>
4994 * dbus/dbus-message.c (dbus_message_append_fields):
4995 Add some more doc comments.
4997 2003-02-01 Havoc Pennington <hp@pobox.com>
4999 * dbus/dbus-break-loader.c (randomly_modify_length): change
5000 a 4-byte value in the message as if it were a length
5002 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): don't set
5003 execute bit on saved files
5005 2003-02-01 Havoc Pennington <hp@pobox.com>
5007 * dbus/dbus-break-loader.c (main): new program to find messages
5008 that break the loader.
5010 * dbus/dbus-sysdeps.c (_dbus_string_append_uint): new function
5011 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): new function
5013 * dbus/dbus-string.c (_dbus_string_set_byte): new
5015 2003-01-31 Havoc Pennington <hp@pobox.com>
5017 * dbus/dbus-message.c: refactor the test code to be more general,
5018 in preparation for writing a "randomly permute test cases to
5019 try to break the loader" program.
5021 2003-01-31 Havoc Pennington <hp@pobox.com>
5023 * doc/dbus-specification.sgml: work on the specification
5025 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): check
5026 the protocol version of the message.
5028 * dbus/dbus-protocol.h: drop special _REPLY names, the spec
5029 no longer specifies that.
5030 (DBUS_SERVICE_REPLY_SERVICE_EXISTS): fix flags (1/2/4/8 not
5033 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos): add missing
5034 "break" for DBUS_TYPE_NIL, remove @todo
5036 2003-01-31 Havoc Pennington <hp@pobox.com>
5038 * dbus/dbus-message.c (dbus_message_set_is_error_reply): rename
5039 just set_is_error/get_is_error as this is a commonly-used
5040 function, and write docs.
5042 2003-01-31 Anders Carlsson <andersca@codefactory.se>
5044 * dbus/dbus-address.c: (dbus_address_entry_free):
5045 Free key and value lists.
5047 * dbus/dbus-internals.c: (_dbus_type_to_string):
5048 Add the types we didn't have.
5050 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
5051 (_dbus_marshal_validate_arg):
5054 * dbus/dbus-message.c: (dbus_message_set_sender):
5055 Remove todo about being able to set sender to NULL.
5057 (dbus_message_set_is_error_reply),
5058 (dbus_message_get_is_error_reply):
5059 * dbus/dbus-message.h:
5062 * dbus/dbus-protocol.h:
5063 Add error reply flag.
5065 * test/data/valid-messages/opposite-endian.message:
5066 Add NIL type to test.
5068 2003-01-31 Havoc Pennington <hp@pobox.com>
5070 * doc/dbus-specification.sgml: fully specify the header. Add
5071 flags and major protocol version, and change header/body len to
5074 * dbus/dbus-message-builder.c (append_saved_length): append length
5077 * dbus/dbus-message.c (dbus_message_create_header): change header
5078 length and body length to unsigned. Add the new fields from the
5080 (_dbus_message_loader_return_buffer): unsigned header/body len
5082 2003-01-30 Havoc Pennington <hp@pobox.com>
5084 * dbus/dbus-auth.c: rework to use only REJECTED, no
5087 * doc/dbus-sasl-profile.txt: drop MECHANISMS and just
5088 use REJECTED, suggested by Mark McLoughlin
5090 2003-01-30 Havoc Pennington <hp@pobox.com>
5092 * dbus/dbus-server.c (dbus_server_listen): @todo about how we need
5093 a better way to report errors here. e.g. "unix address lacks
5094 path" or something. also "no such file" when the path doesn't
5097 * dbus/dbus-address.c (dbus_address_entries_free): add @todo about
5099 (dbus_parse_address): add @todo about documenting address format,
5100 and allowing , and ; to be escaped
5102 2003-01-30 Anders Carlsson <andersca@codefactory.se>
5105 Add dbus-address.[ch]
5107 * dbus/dbus-address.c: (dbus_address_entry_free),
5108 (dbus_address_entries_free), (create_entry),
5109 (dbus_address_entry_get_method), (dbus_address_entry_get_value),
5110 (dbus_parse_address), (_dbus_address_test):
5111 * dbus/dbus-address.h:
5112 New files for dealing with address parsing.
5114 * dbus/dbus-connection.c:
5115 Document timeout functions.
5117 * dbus/dbus-message.c:
5118 Document dbus_message_new_from_message.
5120 * dbus/dbus-server-debug.c:
5123 * dbus/dbus-server.c: (dbus_server_listen):
5124 Parse address and use correct server implementation.
5126 * dbus/dbus-string.c: (_dbus_string_find_to), (_dbus_string_test):
5127 * dbus/dbus-string.h:
5128 New function with test.
5130 * dbus/dbus-test.c: (dbus_internal_symbol_do_not_use_run_tests):
5134 * dbus/dbus-transport-debug.c:
5137 * dbus/dbus-transport.c: (_dbus_transport_open):
5138 Parse address and use correct transport implementation.
5140 2003-01-30 Havoc Pennington <hp@pobox.com>
5142 * dbus/dbus-message.c: use message->byte_order instead of
5143 DBUS_COMPILER_BYTE_ORDER throughout.
5144 (dbus_message_create_header): pad header to align the
5145 start of the body of the message to 8-byte boundary
5147 * dbus/dbus-marshal.h: make all the demarshalers take const
5148 DBusString arguments.
5150 * dbus/dbus-message.c (_dbus_message_loader_return_buffer):
5151 validate message args here, so we don't have to do slow validation
5152 later, and so we catch bad messages as they are incoming. Also add
5153 better checks on header_len and body_len. Also fill in
5156 * dbus/dbus-string.c (_dbus_string_validate_utf8): new (not
5157 implemented properly)
5158 (_dbus_string_validate_nul): new function to check all-nul
5160 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): rename
5161 get_arg_end_pos and remove all validation
5162 (_dbus_marshal_validate_arg): actually do validation here.
5164 2003-01-29 Havoc Pennington <hp@pobox.com>
5166 * dbus/dbus-message.c (check_message_handling): fix assertion
5167 failure on set_client_serial
5169 2003-01-28 Havoc Pennington <hp@pobox.com>
5171 * dbus/dbus-server-debug.c: Add doc section comments
5173 * dbus/dbus-transport-debug.c: add doc section comments
5175 2003-01-28 Havoc Pennington <hp@redhat.com>
5177 * dbus/dbus-string.c (_dbus_string_base64_decode): append bytes in
5178 the reverse order from how I had it
5179 (_dbus_string_base64_encode): reverse encoding order. I was
5180 basically byteswapping everything during encoding.
5182 2003-01-28 Anders Carlsson <andersca@codefactory.se>
5184 * dbus/dbus-connection-internal.h:
5185 * dbus/dbus-connection.c: (_dbus_connection_add_timeout),
5186 (_dbus_connection_remove_timeout):
5187 Add functions for adding and removing timeouts.
5189 * dbus/dbus-message.c: (dbus_message_new_from_message):
5190 Add new function that takes a message and creates an exact
5191 copy of it, but with the refcount set to 1.
5192 (check_message_handling):
5195 * dbus/dbus-server-protected.h:
5196 * dbus/dbus-server.c: (_dbus_server_init_base),
5197 (_dbus_server_finalize_base), (_dbus_server_add_timeout),
5198 (dbus_server_set_timeout_functions):
5199 (_dbus_server_remove_timeout):
5200 New functions so that a server can add and remove timeouts.
5202 (dbus_server_listen):
5203 Add commented out call to dbus_server_debug_new.
5205 * dbus/dbus-timeout.c: (_dbus_timeout_new):
5206 Actually set the handler, doh.
5208 * dbus/dbus-transport.c: (_dbus_transport_open):
5209 Add commented out call to dbus_transport_debug_client_new.
5212 Add dbus-transport-debug.[ch] and dbus-server-debug.[ch]
5214 2003-01-28 Havoc Pennington <hp@pobox.com>
5216 * dbus/dbus-message.c (check_message_handling): function to check
5217 on the loaded message, iterates over it etc.
5219 2003-01-28 Havoc Pennington <hp@pobox.com>
5221 * test/Makefile.am (dist-hook): fix make distdir
5223 * dbus/Makefile.am (TESTS_ENVIRONMENT): fix make check
5225 2003-01-27 Havoc Pennington <hp@pobox.com>
5227 * dbus/dbus-mempool.c (time_for_size): replace printf with
5230 * dbus/dbus-message-builder.c (_dbus_message_data_load): allow
5231 empty lines; fix the SAVE_LENGTH stuff to be
5232 START_LENGTH/END_LENGTH so it actually works; couple other
5235 * test/Makefile.am (dist-hook): add dist-hook for .message files
5237 * dbus/dbus-string.c (DBUS_STRING_COPY_PREAMBLE): source of a copy
5238 can be constant or locked.
5239 (_dbus_string_free): allow freeing a const string as
5242 * dbus/dbus-sysdeps.c (_dbus_concat_dir_and_file): utility
5244 * dbus/dbus-test-main.c (main): take an argument which is the
5245 directory containing test data
5247 * dbus/dbus-message.c (_dbus_message_test): pass a test_data_dir
5248 argument to this and load all the messages in test/data/
5249 checking that they can be loaded or not loaded as appropriate.
5251 2003-01-27 Anders Carlsson <andersca@codefactory.se>
5253 * bus/dispatch.c: (bus_dispatch_message_handler):
5254 Dispatch messages sent to services.
5256 * bus/driver.c: (bus_driver_send_service_deleted),
5257 (bus_driver_send_service_created), (bus_driver_send_service_lost),
5258 (bus_driver_send_service_acquired):
5259 Add helper functions for sending service related messages.
5261 (bus_driver_send_welcome_message):
5262 Send HELLO_REPLY instead of WELCOME.
5264 (bus_driver_handle_list_services):
5265 Send LIST_SERVICES_REPLY instead of SERVICES.
5267 (bus_driver_handle_own_service),
5268 (bus_driver_handle_service_exists):
5269 New message handlers.
5271 (bus_driver_handle_message):
5272 Invoke new message handlers.
5274 (bus_driver_remove_connection):
5275 Don't remove any services here since that's done automatically
5276 by bus_service_remove_owner now.
5279 New function signatures.
5281 * bus/services.c: (bus_service_add_owner):
5282 Send ServiceAcquired message if we're the only primary owner.
5284 (bus_service_remove_owner):
5285 Send ServiceAcquired/ServiceLost messages.
5287 (bus_service_set_prohibit_replacement),
5288 (bus_service_get_prohibit_replacement):
5289 Functions for setting prohibit replacement.
5291 (bus_service_has_owner):
5292 New function that checks if a connection is in the owner queue of
5296 Add new function signatures.
5298 * dbus/dbus-list.c: (_dbus_list_test):
5299 Add tests for _dbus_list_remove_last and traversing the list backwards.
5302 Fix a typo in _dbus_list_get_prev_link, if we're at the first element we can't
5303 go any further, so return NULL then.
5305 * dbus/dbus-protocol.h:
5306 Add new messages, service flags and service replies.
5308 2003-01-26 Havoc Pennington <hp@pobox.com>
5310 * dbus/dbus-message-builder.c: implement, completely untested.
5312 * test/data/*: add data to be used in testing.
5313 ".message" files are our simple loadable text format.
5314 ".message-raw" will be binary dumps of messages.
5316 * dbus/dbus-string.c (_dbus_string_starts_with_c_str): new
5318 2003-01-26 Havoc Pennington <hp@pobox.com>
5320 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): new function
5322 * dbus/dbus-errors.c (dbus_result_to_string): add
5325 * dbus/dbus-message-builder.c: new file, will contain code to load
5326 up messages from files. Not implemented yet.
5328 2003-01-26 Havoc Pennington <hp@pobox.com>
5330 * dbus/dbus-message.c (dbus_message_set_sender): support deleting
5331 the sender by setting to NULL
5333 2003-01-26 Havoc Pennington <hp@pobox.com>
5335 The unit tests pass, but otherwise untested. If it breaks, the
5336 tests should have been better. ;-)
5338 * bus/driver.c (bus_driver_handle_hello): return if we disconnect
5341 * dbus/dbus-message.c: redo everything so we maintain
5342 message->header as the only copy of the various fields.
5343 This avoids the possibility of out-of-memory in some cases,
5344 for example dbus_message_lock() can't run out of memory anymore,
5345 and avoids extra copying. Figured I may as well go ahead and do
5346 this since it was busted for dbus_message_lock to not return
5347 failure on OOM, and dbus_message_write_header was totally
5348 unchecked for OOM. Also fixed some random other bugs.
5350 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): verify
5351 that strings are nul-terminated. Also, end_pos can be equal
5352 to string length just not greater than, I think.
5353 (_dbus_marshal_set_int32): new function
5354 (_dbus_marshal_set_uint32): new function
5355 (_dbus_marshal_set_string): new function
5357 * dbus/dbus-connection.c (_dbus_connection_new_for_transport): fix
5358 a warning, init timeout_list to NULL
5359 (dbus_connection_send_message): don't use uninitialized variable
5362 * dbus/dbus-string.c (_dbus_string_replace_len): new function
5364 2003-01-26 Anders Carlsson <andersca@codefactory.se>
5366 * bus/driver.c: (bus_driver_handle_hello),
5367 (bus_driver_send_welcome_message):
5370 2003-01-26 Anders Carlsson <andersca@codefactory.se>
5372 * dbus/dbus-auth.c: (process_auth), (_dbus_auth_unref):
5373 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
5374 (dbus_connection_unref):
5375 * dbus/dbus-marshal.c: (_dbus_marshal_test):
5376 * dbus/dbus-message.c: (dbus_message_unref),
5379 (dbus_message_get_fields):
5380 Remove debugging printout.
5382 (_dbus_message_loader_return_buffer):
5383 Don't store the header string.
5385 (_dbus_message_test):
5388 2003-01-26 Richard Hult <rhult@codefactory.se>
5390 * glib/dbus-gmain.c (dbus_connection_dispatch): Traverse a copy of
5391 the file descriptor list, since it can change under us.
5393 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5395 * glib/dbus-gmain.c: (dbus_connection_prepare),
5396 (dbus_connection_check), (dbus_connection_dispatch), (add_watch),
5397 (remove_watch), (dbus_connection_hookup_with_g_main):
5398 Rewrite the glib handling to use its own GSource instead of a
5399 GIOChannel so we can catch messages put in the queue while waiting
5402 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5405 * bus/connection.c: (connection_disconnect_handler),
5406 (connection_watch_callback), (bus_connection_setup):
5407 * bus/dispatch.c: (send_one_message),
5408 (bus_dispatch_broadcast_message), (bus_dispatch_message_handler),
5409 (bus_dispatch_add_connection), (bus_dispatch_remove_connection):
5411 * bus/driver.c: (bus_driver_send_service_deleted),
5412 (bus_driver_send_service_created), (bus_driver_handle_hello),
5413 (bus_driver_send_welcome_message),
5414 (bus_driver_handle_list_services), (bus_driver_remove_connection),
5415 (bus_driver_handle_message):
5417 Refactor code, put the message dispatching in its own file. Use
5418 _DBUS_HANDLE_OOM. Also send ServiceDeleted messages when a client
5421 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5423 * dbus/dbus-internals.h:
5424 Add _DBUS_HANDLE_OOM macro, it doesn't do anything currently.
5426 * dbus/dbus-message.c: (dbus_message_get_sender):
5427 * dbus/dbus-message.h:
5428 Implement dbus_message_get_sender.
5430 * dbus/dbus-protocol.h:
5431 Add message and service defines.
5433 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5435 * dbus/dbus-connection.c: (dbus_connection_send_message):
5436 * dbus/dbus-message-internal.h:
5437 * dbus/dbus-message.c: (_dbus_message_get_client_serial),
5438 (dbus_message_write_header):
5439 Remove _dbus_messag_unlock and don't set the client serial on a
5440 message if one already exists.
5442 2003-01-24 Havoc Pennington <hp@pobox.com>
5444 * dbus/dbus-list.c (alloc_link): put a thread lock on the global
5447 * bus/driver.c (bus_driver_handle_list_services): fix a leak
5450 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5452 * dbus/dbus-list.c: (alloc_link), (free_link):
5453 Use a memory pool for the links.
5455 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5457 * bus/connection.c: (bus_connection_foreach):
5459 Add new bus_connection_foreach function.
5461 * bus/driver.c: (send_one_message), (bus_driver_broadcast_message):
5462 Add function that broadcasts a message to all clients.
5464 (bus_driver_send_service_created), (bus_driver_handle_hello),
5465 (bus_driver_send_welcome_message),
5466 (bus_driver_handle_list_services), (bus_driver_message_handler):
5467 Implement functions that take care of listing services, and notifying
5468 clients when new services are created.
5470 * bus/services.c: (bus_services_list):
5472 Add new function that returns an array of strings with the currently
5473 registered services.
5476 * glib/dbus-gmain.c:
5477 Update copyright year.
5479 2003-01-25 Anders Carlsson <andersca@codefactory.se>
5481 * dbus/dbus-connection.c: (dbus_connection_send_message):
5482 Unlock the message in case it was sent earlier.
5484 (dbus_connection_send_message_with_reply_and_block):
5485 Remove the reply message from the list.
5487 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
5488 Set array_len and new_pos correctly.
5490 (_dbus_marshal_test):
5491 Remove debug output.
5493 * dbus/dbus-message-internal.h:
5494 * dbus/dbus-message.c: (_dbus_message_get_reply_serial):
5495 New function that returns the reply serial.
5497 (_dbus_message_unlock):
5498 New function that unlocks a message and resets its header.
5500 (dbus_message_append_string_array),
5501 (dbus_message_get_fields_valist),
5502 (dbus_message_iter_get_field_type),
5503 (dbus_message_iter_get_string_array),
5504 (dbus_message_get_fields),
5505 (dbus_message_append_fields_valist):
5506 Handle string arrays.
5508 (dbus_message_set_sender):
5509 Make this function public since the bus daemon needs it.
5511 (decode_header_data):
5512 Set the reply serial to -1 initially.
5514 * dbus/dbus-message.h:
5515 Add dbus_message_set_sender.
5517 2003-01-24 Havoc Pennington <hp@pobox.com>
5519 * doc/dbus-specification.sgml: add some stuff
5521 2003-01-22 Havoc Pennington <hp@pobox.com>
5523 * doc/dbus-specification.sgml: Start to document the protocol.
5525 2003-01-22 Havoc Pennington <hp@pobox.com>
5527 * dbus/dbus-connection.c
5528 (dbus_connection_send_message_with_reply_and_block): add some @todo
5530 * bus/driver.c (bus_driver_add_connection): add a FIXME about memleak
5532 2003-01-21 Havoc Pennington <hp@pobox.com>
5534 (patch untested because can't compile)
5536 * bus/driver.c (create_unique_client_name): make this function
5537 never recycle client names. Also, caller should initialize
5540 * dbus/dbus-sysdeps.c (_dbus_get_current_time): new function
5542 2003-01-21 Anders Carlsson <andersca@codefactory.se>
5544 * dbus/dbus-marshal.c: (_dbus_marshal_double),
5545 (_dbus_marshal_int32), (_dbus_marshal_uint32),
5546 (_dbus_marshal_int32_array), (_dbus_marshal_uint32_array),
5547 (_dbus_marshal_double_array), (_dbus_marshal_string_array),
5548 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
5549 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
5550 (_dbus_marshal_get_field_end_pos), (_dbus_marshal_test):
5551 * dbus/dbus-marshal.h:
5552 * dbus/dbus-protocol.h:
5553 Add support for marshalling and demarshalling integer, double
5556 2003-01-21 Anders Carlsson <andersca@codefactory.se>
5561 * bus/connection.c: (connection_disconnect_handler):
5562 Remove the connection from the bus driver's list.
5564 (connection_watch_callback): Dispatch messages.
5566 (free_connection_data): Free connection name.
5568 (bus_connection_setup): Add connection to the bus driver's list.
5569 (bus_connection_remove_owned_service):
5570 (bus_connection_set_name), (bus_connection_get_name):
5571 Add functions for setting and getting the connection's name.
5574 Add function headers.
5576 * bus/driver.c: (create_unique_client_name),
5577 (bus_driver_handle_hello_message),
5578 (bus_driver_send_welcome_message), (bus_driver_message_handler),
5579 (bus_driver_add_connection), (bus_driver_remove_connection):
5582 * bus/services.c: (bus_service_free):
5584 New file that handles communication and registreation with the bus
5587 2003-01-21 Anders Carlsson <andersca@codefactory.se>
5589 * dbus/dbus-connection.c: (dbus_connection_send_message):
5590 Add a new client_serial parameter.
5592 (dbus_connection_send_message_with_reply):
5593 Remove a @todo since we've implemented the blocking function.
5595 (dbus_connection_send_message_with_reply_and_block):
5596 New function that sends a message and waits for a reply and
5597 then returns the reply.
5599 * dbus/dbus-connection.h:
5602 * dbus/dbus-errors.c: (dbus_result_to_string):
5603 * dbus/dbus-errors.h:
5604 Add new DBUS_RESULT.
5606 * dbus/dbus-message-internal.h:
5607 * dbus/dbus-message.c: (_dbus_message_get_reply_serial),
5608 (_dbus_message_set_sender), (dbus_message_write_header),
5609 (dbus_message_new_reply), (decode_header_data),
5610 (_dbus_message_loader_return_buffer), (_dbus_message_test):
5611 * dbus/dbus-message.h:
5612 Add new functions that set the reply serial and sender.
5613 Also marshal and demarshal them correctly and add test.
5615 * dbus/dbus-protocol.h:
5616 Add new DBUS_MESSAGE_TYPE_SENDER.
5619 * glib/dbus-gmain.c: (watch_callback), (free_callback_data),
5620 (add_watch), (remove_watch), (add_timeout), (remove_timeout),
5621 (dbus_connection_hookup_with_g_main):
5622 * glib/test-dbus-glib.c: (main):
5623 Rewrite to use GIOChannel and remove the GSource crack.
5625 * test/echo-client.c: (main):
5626 * test/watch.c: (check_messages):
5627 Update for changed APIs
5629 2003-01-19 Anders Carlsson <andersca@codefactory.se>
5631 * dbus/Makefile.am: Add dbus-timeout.[cħ]
5633 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport):
5634 Create a DBusTimeoutList.
5635 (dbus_connection_set_timeout_functions): Add new function to
5636 set timeout callbacks
5638 * dbus/dbus-connection.h: Add public DBusTimeout API.
5640 * dbus/dbus-message.c: (dbus_message_get_service):
5641 * dbus/dbus-message.h: New function.
5643 * dbus/dbus-server.c: Fix small doc typo.
5645 * dbus/dbus-timeout.[ch]: New files for mainloop timeouts.
5647 2003-01-19 Anders Carlsson <andersca@codefactory.se>
5649 * dbus/dbus-string.c (_dbus_string_move_len): Don't delete all
5650 of the string, just as long as specified.
5652 2003-01-19 Havoc Pennington <hp@pobox.com>
5654 * dbus/dbus-connection.c (dbus_connection_get_is_authenticated):
5657 * dbus/dbus-server.c (dbus_server_set_max_connections)
5658 (dbus_server_get_max_connections, dbus_server_get_n_connections):
5659 keep track of current number of connections, and add API for
5660 setting a max (but haven't implemented enforcing the max yet)
5662 2003-01-18 Havoc Pennington <hp@pobox.com>
5664 * dbus/dbus-transport-unix.c (unix_do_iteration): only do the
5665 reading/writing if read_watch != NULL or write_watch != NULL.
5667 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): fix
5668 the message loader code to actually load message->header and
5669 message->body into the newly-created message.
5671 * dbus/dbus-transport-unix.c (check_write_watch): fix a mem leak
5674 * dbus/dbus-connection.c (dbus_connection_set_max_message_size)
5675 (dbus_connection_get_max_message_size)
5676 (dbus_connection_set_max_live_messages_size)
5677 (dbus_connection_get_max_live_messages_size): implement some
5678 resource limitation functions
5680 * dbus/dbus-resources.c: new file implementing some of the
5681 resource limits stuff
5683 * dbus/dbus-message.c (dbus_message_iter_get_byte_array): add
5684 missing docs, add @todo to handle OOM etc.
5686 * dbus/dbus-marshal.c (_dbus_demarshal_byte_array): add missing
5689 2003-01-18 Havoc Pennington <hp@pobox.com>
5691 * dbus/dbus-connection.c (dbus_connection_unref): disconnect the
5692 connection if it hasn't been already.
5694 * dbus/dbus-connection.h: kill off the idea of an ErrorFunction,
5695 replace with DisconnectFunction.
5697 2003-01-18 Havoc Pennington <hp@pobox.com>
5699 Building --disable-verbose-mode --disable-asserts --disable-tests
5700 cuts the library from 112K to 45K or so
5702 * configure.in: check for varargs macro support,
5703 add --enable-verbose-mode, --enable-asserts.
5705 * dbus/dbus-internals.h (_dbus_assert): support
5707 (_dbus_verbose): support DBUS_ENABLE_VERBOSE_MODE
5709 2003-01-18 Havoc Pennington <hp@pobox.com>
5711 * dbus/dbus-test.c: include config.h so that tests actually run
5713 * dbus/dbus-string.c: add assertions that stuff is 8-byte aligned,
5714 so the failure mode when that assumption fails will be plenty
5717 2003-01-18 Havoc Pennington <hp@pobox.com>
5719 * configure.in: default --enable-tests to $USE_MAINTAINER_MODE
5721 * dbus/Makefile.am: fix it up so dubs-test-main.c is included in
5724 * test/Makefile.am: don't use special variable "TESTS" for echo-*
5725 since we don't want to use those in make check
5727 2003-01-15 Havoc Pennington <hp@redhat.com>
5733 2003-01-15 Havoc Pennington <hp@redhat.com>
5735 * test/Makefile.am: fix so that test source code ends up in the
5736 distribution on make distcheck
5738 2003-01-15 Havoc Pennington <hp@redhat.com>
5744 2003-01-15 Havoc Pennington <hp@redhat.com>
5746 * dbus/dbus-test.c (dbus_internal_symbol_do_not_use_run_tests):
5747 fix build when --disable-tests
5749 * Makefile.am (EXTRA_DIST): put HACKING in here
5751 * HACKING: document procedure for making a tarball release.
5753 2003-01-14 Anders Carlsson <andersca@codefactory.se>
5755 * bus/connection.c: (connection_error_handler),
5756 (bus_connection_setup):
5757 * bus/main.c: (main):
5758 Make sure that the DBusConnectionData struct is NULLed
5759 out to prevent a segfault.
5761 * dbus/dbus-errors.c: (dbus_result_to_string):
5762 * dbus/dbus-errors.h:
5763 * dbus/dbus-message.c: (dbus_message_get_fields),
5764 (dbus_message_get_fields_valist), (_dbus_message_test):
5765 * dbus/dbus-message.h:
5766 Make dbus_message_get_fields return a result code so we can
5767 track invalid fields as well as oom.
5769 2003-01-11 Havoc Pennington <hp@pobox.com>
5771 * configure.in: change --enable-test/--enable-ansi action-if-given
5772 to enable_foo=$enableval instead of enable_foo=yes
5774 2003-01-08 Havoc Pennington <hp@pobox.com>
5776 * dbus/dbus-string.c (_dbus_string_align_length): new function
5778 * dbus/dbus-test-main.c: move main() for test app here
5780 (dbus_internal_symbol_do_not_use_run_tests): we have to export a
5781 symbol to run tests, because dbus-test isn't in the main
5784 Code review nitpicks.
5786 * dbus/dbus-message.c (dbus_message_write_header): add newlines
5787 for people with narrow emacs ;-). Assert client_serial was filled
5788 in. Assert message->name != NULL.
5789 (dbus_message_append_fields): have "first_field_type" arg separate
5790 from va list, needed for C++ binding that also uses varargs IIRC
5791 and helps with type safety
5792 (dbus_message_new): add @todo about using DBusString to store
5793 service/name internally
5794 (dbus_message_new): don't leak ->service and ->name on OOM later
5796 (dbus_message_unref): free the service name
5797 (dbus_message_get_fields): same change to varargs
5798 i.e. first_field_type
5799 (_dbus_message_loader_return_buffer): assert that the message data
5800 is aligned (if not it's a bug in our code). Put in verbose griping
5801 about why we set corrupted = TRUE.
5802 (decode_header_data): add FIXME that char* is evil. Was going to
5803 add FIXME about evil locale-specific string.h strncmp, but just
5804 switched to wacky string-as-uint32 optimization. Move check for
5805 "no room for field name" above get_const_data_len() to avoid
5806 assertion failure in get_const_data_len if we have trailing 2
5807 bytes or the like. Check for service and name fields being
5808 provided twice. Don't leak service/name on error. Require field
5809 names to be aligned to 4 bytes.
5811 * dbus/dbus-marshal.c: move byte swap stuff to header
5812 (_dbus_pack_int32): uscore-prefix
5813 (_dbus_unpack_int32): uscore-prefix
5814 (_dbus_unpack_uint32): export
5815 (_dbus_demarshal_string): add @todo complaining about use of
5817 (_dbus_marshal_get_field_end_pos): add @todo about bad error
5818 handling allowing corrupt data to go unchecked
5820 2003-01-08 Havoc Pennington <hp@redhat.com>
5822 * dbus/dbus-transport-unix.c (unix_do_iteration): add read/write
5823 to the select() as needed for authentication. (should be using
5824 _dbus_poll() not select, but for another day)
5826 * dbus/dbus.h: include dbus/dbus-protocol.h
5828 2003-01-08 Anders Carlsson <andersca@codefactory.se>
5830 * dbus/Makefile.am (dbusinclude_HEADERS): Install
5833 2003-01-08 Anders Carlsson <andersca@codefactory.se>
5835 * dbus/dbus-internals.c: (_dbus_type_to_string):
5836 New function that returns a string describing a type.
5838 * dbus/dbus-marshal.c: (_dbus_demarshal_byte_array):
5839 * dbus/dbus-marshal.h:
5840 * dbus/dbus-message.c: (dbus_message_get_fields_valist),
5841 (dbus_message_iter_get_field_type), (dbus_message_iter_get_double),
5842 (dbus_message_iter_get_byte_array):
5843 * dbus/dbus-message.h:
5844 Add new convenience functions for appending and getting message fields.
5845 Also add demarshalling routines for byte arrays.
5847 2003-01-07 Anders Carlsson <andersca@codefactory.se>
5849 * dbus/dbus-connection-internal.h:
5850 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
5851 (_dbus_connection_get_next_client_serial),
5852 (dbus_connection_send_message):
5853 * dbus/dbus-internals.h:
5854 * dbus/dbus-marshal.c: (unpack_uint32), (dbus_unpack_int32),
5855 (dbus_pack_int32), (_dbus_marshal_double), (_dbus_marshal_int32),
5856 (_dbus_marshal_uint32), (_dbus_demarshal_double),
5857 (_dbus_demarshal_int32), (_dbus_demarshal_uint32),
5858 (_dbus_demarshal_string), (_dbus_marshal_get_field_end_pos),
5859 (_dbus_verbose_bytes), (_dbus_marshal_test):
5860 * dbus/dbus-marshal.h:
5861 * dbus/dbus-message-internal.h:
5862 * dbus/dbus-message.c: (_dbus_message_set_client_serial),
5863 (dbus_message_write_header), (_dbus_message_lock),
5864 (dbus_message_new), (dbus_message_ref), (dbus_message_unref),
5865 (dbus_message_get_name), (dbus_message_append_int32),
5866 (dbus_message_append_uint32), (dbus_message_append_double),
5867 (dbus_message_append_string), (dbus_message_append_byte_array),
5868 (dbus_message_get_fields_iter), (dbus_message_iter_ref),
5869 (dbus_message_iter_unref), (dbus_message_iter_has_next),
5870 (dbus_message_iter_next), (dbus_message_iter_get_field_type),
5871 (dbus_message_iter_get_string), (dbus_message_iter_get_int32),
5872 (dbus_message_iter_get_uint32), (dbus_message_iter_get_double),
5873 (decode_header_data), (_dbus_message_loader_return_buffer),
5874 (message_iter_test), (_dbus_message_test):
5875 * dbus/dbus-message.h:
5876 * dbus/dbus-protocol.h:
5877 * dbus/dbus-test.c: (main):
5879 * glib/test-dbus-glib.c: (message_handler), (main):
5880 * test/echo-client.c: (main):
5881 * test/watch.c: (check_messages):
5882 Make messages sendable and receivable for real.
5884 2003-01-07 Anders Carlsson <andersca@codefactory.se>
5886 * dbus/dbus-marshal.c: (_dbus_marshal_double),
5887 (_dbus_marshal_string), (_dbus_marshal_byte_array):
5888 * dbus/dbus-message.c: (dbus_message_append_int32),
5889 (dbus_message_append_uint32), (dbus_message_append_double),
5890 (dbus_message_append_string), (dbus_message_append_byte_array):
5891 Handle OOM restoration.
5893 2003-01-07 Anders Carlsson <andersca@codefactory.se>
5895 * dbus/dbus-marshal.c: (_dbus_marshal_string),
5896 (_dbus_demarshal_string), (_dbus_marshal_test):
5897 * dbus/dbus-marshal.h:
5898 * dbus/dbus-message.c: (dbus_message_get_name),
5899 Document these functions.
5901 (dbus_message_append_int32), (dbus_message_append_uint32),
5902 (dbus_message_append_double), (dbus_message_append_string),
5903 (dbus_message_append_byte_array):
5904 * dbus/dbus-message.h:
5905 Add functions for adding message fields of different types.
5907 * dbus/dbus-protocol.h:
5908 Add the different types.
5910 2003-01-05 Havoc Pennington <hp@pobox.com>
5912 * bus/connection.c: implement routines for handling connections,
5913 first thing is keeping a list of owned services on each connection
5914 and setting up watches etc.
5916 * bus/services.c: implement a mapping from service names to lists
5919 * dbus/dbus-hash.c: add DBUS_HASH_POINTER
5921 * dbus/dbus-threads.c (dbus_static_mutex_lock): add functions
5922 to use static mutexes for global data
5924 * dbus/dbus-connection.c (dbus_connection_set_data): add new
5925 collection of functions to set/get application-specific data
5926 on the DBusConnection.
5928 2003-01-04 Havoc Pennington <hp@pobox.com>
5930 * dbus/dbus-sysdeps.c (_dbus_sleep_milliseconds): new function
5931 (_dbus_poll): new function
5933 * dbus/dbus-internals.h (_DBUS_STRUCT_OFFSET): new macro
5936 * bus/loop.c: initial code for the daemon main loop
5938 2003-01-04 Havoc Pennington <hp@pobox.com>
5940 * test/watch.c (error_handler): make it safe if the error handler
5941 is called multiple times (if we s/error handler/disconnect
5942 handler/ we should just guarantee it's called only once)
5944 * dbus/dbus-transport.c (_dbus_transport_disconnect): call the
5945 error handler on disconnect (it's quite possible we should
5946 just change the error handler to a "disconnect handler," I'm
5947 not sure we have any other meaningful errors)
5949 * configure.in: check for getpwnam_r
5951 * dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
5952 dbus/dbus-auth.c: add credentials support, add EXTERNAL auth
5953 mechanism as in SASL spec, using socket credentials
5955 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): new function
5956 (_dbus_send_credentials_unix_socket): new function
5958 * dbus/dbus-sysdeps.c (_dbus_accept_unix_socket): rename just
5960 (_dbus_write): only check errno if <0 returned
5961 (_dbus_write_two): ditto
5963 2003-01-02 Anders Carlsson <andersca@codefactory.se>
5965 * dbus/dbus-marshal.c: (_dbus_marshal_utf8_string),
5966 (_dbus_marshal_byte_array), (_dbus_demarshal_utf8_string),
5967 (_dbus_marshal_test):
5968 * dbus/dbus-marshal.h:
5969 Add _dbus_marshal_byte_array and rename _dbus_marshal_string
5970 to _dbus_marshal_utf8_string. Also fix some tests.
5972 2002-12-28 Harri Porten <porten@kde.org>
5974 * configure.in: added check for C++ compiler and a very cheesy
5975 check for the Qt integration
5977 * Makefile.am (SUBDIRS): compile qt subdir if support is enabled
5979 * qt/Makefile.am: added
5981 * qt/.cvsignore: added
5983 * qt/dbus-qthread.cc, qt/dbus-qthread.cpp: renamed former to
5984 latter, added #ifdef QT_THREAD_SUPPORT guard.
5986 * dbus/Makefile.am: added missing headers for make dist
5988 2002-12-28 Kristian Rietveld <kris@gtk.org>
5990 * dbus/Makefile.am: fixup export-symbols-regex.
5992 2002-12-27 Anders Carlsson <andersca@codefactory.se>
5994 * acinclude.m4: Add this file and put the
5995 PKG_CHECK_MODULE macro in it.
5997 2002-12-27 Anders Carlsson <andersca@codefactory.se>
5999 * dbus/dbus-marshal.c: (_dbus_marshal_string),
6000 (_dbus_demarshal_double), (_dbus_demarshal_int32),
6001 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
6002 (_dbus_marshal_test):
6003 Make the demarshalling routines align the pos argument.
6004 Add string marshalling tests and fix the obvious bugs
6007 2002-12-26 Havoc Pennington <hp@pobox.com>
6009 * dbus/dbus-auth.c: fixes fixes fixes
6011 * dbus/dbus-transport-unix.c: wire up support for
6012 encoding/decoding data on the wire
6014 * dbus/dbus-auth.c (_dbus_auth_encode_data)
6015 (_dbus_auth_decode_data): append to target string
6016 instead of nuking it.
6018 2002-12-26 Havoc Pennington <hp@pobox.com>
6020 * dbus/dbus-marshal.h (DBUS_COMPILER_BYTE_ORDER): #ifdef
6021 WORDS_BIGENDIAN then compiler byte order is DBUS_BIG_ENDIAN,
6024 * dbus/dbus-marshal.c: Add macros to do int swapping in-place and
6025 avoid swap_bytes() overhead (ignoring possible assembly stuff for
6026 now). Main point is because I wanted unpack_uint32 to implement
6028 (_dbus_verbose_bytes): new function
6030 * dbus/dbus-string.c (_dbus_string_validate_ascii): new function
6032 * dbus/dbus-message.c (_dbus_message_loader_get_is_corrupted): add
6033 mechanism to handle a corrupt message stream
6034 (_dbus_message_loader_new): fix preallocation to only prealloc,
6037 * dbus/dbus-string.c (_dbus_string_skip_blank): fix this function
6038 (_dbus_string_test): enhance tests for copy/move and fix the
6041 * dbus/dbus-transport-unix.c: Hold references in more places to
6042 avoid reentrancy problems
6044 * dbus/dbus-transport.c: ditto
6046 * dbus/dbus-connection.c (dbus_connection_dispatch_message): don't
6047 leak reference count in no-message case
6049 * test/watch.c (do_mainloop): handle adding/removing watches
6050 during iteration over the watches. Also, ref the connection/server
6051 stored on a watch, so we don't try to mangle a destroyed one.
6053 * dbus/dbus-transport-unix.c (do_authentication): perform
6056 * dbus/dbus-auth.c (get_state): add a state
6057 AUTHENTICATED_WITH_UNUSED_BYTES and return it if required
6058 (_dbus_auth_get_unused_bytes): append the unused bytes
6059 to the passed in string, rather than prepend
6061 * dbus/dbus-transport.c (_dbus_transport_init_base): create
6062 the auth conversation DBusAuth
6064 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd)
6065 (_dbus_transport_new_for_domain_socket): when creating a
6066 transport, pass in whether it's a client-side or server-side
6067 transport so we know which DBusAuth to create
6069 2002-12-03 Havoc Pennington <hp@pobox.com>
6071 * dbus/dbus-transport-unix.c (unix_finalize): finalize base
6072 _after_ finalizing the derived members
6073 (unix_connection_set): unref watch if we fail to add it
6075 * dbus/dbus-connection.c (dbus_connection_unref): delete the
6076 transport first, so that the connection owned by the
6077 transport will be valid as the transport finalizes.
6079 * dbus/dbus-transport-unix.c (unix_finalize): free the write_watch
6080 if necessary, and remove watches from the connection.
6082 * dbus/dbus-watch.c (_dbus_watch_list_free): improve a comment
6084 2002-12-26 Anders Carlsson <andersca@codefactory.se>
6086 * dbus/dbus-marshal.c: (_dbus_marshal_string),
6087 (_dbus_demarshal_double), (_dbus_demarshal_int32),
6088 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
6089 (_dbus_marshal_test):
6090 * dbus/dbus-marshal.h:
6091 Add string marshal functions and have the demarshal functions
6092 return the new position.
6094 2002-12-25 Havoc Pennington <hp@pobox.com>
6096 * doc/dbus-sasl-profile.txt: docs on the authentication protocol,
6097 it is a simple protocol that just maps directly to SASL.
6099 * dbus/dbus-auth.h, dbus/dbus-auth.c: authentication protocol
6100 initial implementation, not actually used yet.
6102 * dbus/dbus-string.c (_dbus_string_find): new function
6103 (_dbus_string_equal): new function
6104 (_dbus_string_base64_encode): new function
6105 (_dbus_string_base64_decode): new function
6107 2002-12-25 Anders Carlsson <andersca@codefactory.se>
6110 * dbus/dbus-marshal.c: (swap_bytes), (_dbus_marshal_double),
6111 (_dbus_marshal_int32), (_dbus_marshal_uint32),
6112 (_dbus_demarshal_double), (_dbus_demarshal_int32),
6113 (_dbus_demarshal_uint32), (_dbus_marshal_test):
6114 * dbus/dbus-marshal.h:
6115 * dbus/dbus-protocol.h:
6116 * dbus/dbus-test.c: (main):
6118 Add un-optimized marshalling/demarshalling routines.
6120 2002-12-25 Harri Porten <porten@kde.org>
6122 * qt/dbus-qt.h: adjusted ctor and getter to KDE/Qt conventions
6124 2002-12-24 Zack Rusin <zack@kde.org>
6126 * qt/dbus-qthread.cc: adding - integrates QMutex into Dbus
6127 * qt/dbus-qt.h: skeleton with two sample implemenatation of the
6130 2002-12-24 Havoc Pennington <hp@pobox.com>
6132 * glib/dbus-gthread.c: fix include
6134 * glib/dbus-glib.h: rename DBusMessageHandler for now.
6135 I think glib API needs to change, though, as you don't
6136 want to use DBusMessageFunction, you want to use the
6137 DBusMessageHandler object. Probably
6138 dbus_connection_open_with_g_main_loop()
6139 and dbus_connection_setup_g_main_loop() or something like that
6140 (but think of better names...) that just create a connection
6141 that has watch/timeout functions etc. already set up.
6143 * dbus/dbus-connection.c
6144 (dbus_connection_send_message_with_reply): new function just to
6145 show how the message handler helps us deal with replies.
6147 * dbus/dbus-list.c (_dbus_list_remove_last): new function
6149 * dbus/dbus-string.c (_dbus_string_test): free a string that
6152 * dbus/dbus-hash.c: use memory pools for the hash entries
6153 (rebuild_table): be more paranoid about overflow, and
6154 shrink table when we can
6155 (_dbus_hash_test): reduce number of sprintfs and write
6156 valid C89. Add tests for case where we grow and then
6157 shrink the hash table.
6159 * dbus/dbus-mempool.h, dbus/dbus-mempool.c: memory pools
6161 * dbus/dbus-connection.c (dbus_connection_register_handler)
6162 (dbus_connection_unregister_handler): new functions
6164 * dbus/dbus-message.c (dbus_message_get_name): new
6166 * dbus/dbus-list.c: fix docs typo
6168 * dbus/dbus-message-handler.h, dbus/dbus-message-handler.c:
6169 an object representing a handler for messages.
6171 2002-12-16 Anders Carlsson <andersca@codefactory.se>
6174 * glib/dbus-gthread.c: (dbus_gthread_init):
6175 Don't use the gdbus prefix for public functions.
6177 2002-12-16 Anders Carlsson <andersca@codefactory.se>
6181 Add GLib checks and fixup .pc files
6185 * glib/dbus-gmain.c: (gdbus_connection_prepare),
6186 (gdbus_connection_check), (gdbus_connection_dispatch),
6187 (gdbus_add_connection_watch), (gdbus_remove_connection_watch),
6188 (dbus_connection_gsource_new):
6189 * glib/dbus-gthread.c: (dbus_gmutex_new), (dbus_gmutex_free),
6190 (dbus_gmutex_lock), (dbus_gmutex_unlock), (dbus_gthread_init):
6191 * glib/test-dbus-glib.c: (message_handler), (main):
6194 2002-12-15 Harri Porten <porten@kde.org>
6196 * autogen.sh: check for libtoolize before attempting to use it
6198 * dbus/dbus-transport-unix.c: include <sys/time.h> for timeval
6201 * .cvsignore: ignore more stamp files
6203 * dbus/dbus-watch.c (_dbus_watch_list_new): fixed doc error
6205 * test/Makefile.am: added -I$(top_srcdir) to be able to compile
6206 without make install.
6208 2002-12-15 Havoc Pennington <hp@pobox.com>
6210 * dbus/dbus-threads.c: add thread stubs that a higher library
6211 layer can fill in. e.g. the GLib wrapper might fill them in with
6212 GThread stuff. We still need to use this thread API to
6213 thread-safe-ize the library.
6215 2002-12-12 Havoc Pennington <hp@pobox.com>
6217 * dbus/dbus-transport-unix.c, dbus/dbus-server-unix.c: use the
6218 below new interfaces and include fewer system headers.
6220 * dbus/dbus-sysdeps.c (_dbus_read): new function
6221 (_dbus_write): new function
6222 (_dbus_write_two): new function
6223 (_dbus_connect_unix_socket): new function
6224 (_dbus_listen_unix_socket): new function
6226 * dbus/dbus-message-internal.h: change interfaces to use
6229 2002-12-11 Havoc Pennington <hp@pobox.com>
6231 * dbus/dbus-types.h: add dbus_unichar
6233 * dbus/dbus-internals.c (_dbus_verbose): use _dbus_getenv
6235 * dbus/dbus-connection.c (dbus_connection_send_message): return
6238 * dbus/dbus-transport.c: include dbus-watch.h
6240 * dbus/dbus-connection.c: include dbus-message-internal.h
6242 * HACKING: add file with coding guidelines stuff.
6244 * dbus/dbus-string.h, dbus/dbus-string.c: Encapsulate all string
6245 handling here, for security purposes (as in vsftpd). Not actually
6246 using this class yet.
6248 * dbus/dbus-sysdeps.h, dbus/dbus-sysdeps.c: Encapsulate all
6249 system/libc usage here, as in vsftpd, for ease of auditing (and
6250 should also simplify portability). Haven't actually moved all the
6251 system/libc usage into here yet.
6253 2002-11-25 Havoc Pennington <hp@pobox.com>
6255 * dbus/dbus-internals.c (_dbus_verbose): fix to not
6256 always print the first verbose message.
6258 2002-11-24 Havoc Pennington <hp@pobox.com>
6260 * test/echo-client.c, test/echo-server.c: cheesy test
6263 * configure.in (AC_CHECK_FUNCS): check for writev
6265 * dbus/dbus-message.c (_dbus_message_get_network_data): new
6268 * dbus/dbus-list.c (_dbus_list_foreach): new function
6270 * dbus/dbus-internals.c (_dbus_verbose): new function
6272 * dbus/dbus-server.c, dbus/dbus-server.h: public object
6273 representing a server that listens for connections.
6275 * dbus/.cvsignore: create
6277 * dbus/dbus-errors.h, dbus/dbus-errors.c:
6278 public API for reporting errors
6280 * dbus/dbus-connection.h, dbus/dbus-connection.c:
6281 public object representing a connection that
6282 sends/receives messages. (Same object used for
6283 both client and server.)
6285 * dbus/dbus-transport.h, dbus/dbus-transport.c:
6286 Basic abstraction for different kinds of stream
6287 that we might read/write messages from.
6289 2002-11-23 Havoc Pennington <hp@pobox.com>
6291 * dbus/dbus-internals.h (_DBUS_INT_MAX): add _DBUS_INT_MIN
6294 * dbus/dbus-test.c (main): add list test, and include
6295 dbus-test.h as intended
6297 * dbus/dbus-hash.c (_dbus_hash_table_remove_string)
6298 (_dbus_hash_table_remove_int): return value indicates
6299 whether the entry existed to remove
6301 * dbus/dbus-list.c: add linked list utility class,
6304 * dbus/dbus-hash.c: add TODO item about shrinking the hash bucket
6307 2002-11-23 Havoc Pennington <hp@pobox.com>
6309 * Doxyfile.in (INCLUDE_FILE_PATTERNS): expand DBUS_BEGIN_DECLS/
6310 DBUS_END_DECLS to nothing, that should fix this once and for all
6312 * Doxyfile.in (JAVADOC_AUTOBRIEF): set to YES
6314 * dbus/dbus-message.c, dbus/dbus-hash.c:
6315 add some missing @brief
6317 2002-11-23 Havoc Pennington <hp@pobox.com>
6319 * dbus/dbus-message.h: put semicolons after DEBUG_BEGIN_DECLS
6320 to avoid confusing Doxygen
6322 * dbus/dbus-hash.c: @} not }@
6324 * dbus/dbus-message.c (struct DBusMessage): split out
6327 2002-11-23 Havoc Pennington <hp@pobox.com>
6329 * configure.in: pile on more warning flags if using gcc
6331 * Doxyfile.in (EXTRACT_STATIC): set to NO, so we don't have
6332 to document static functions
6334 * configure.in: add summary to end of configure so it
6335 looks nice and attractive
6337 * dbus/dbus-hash.c: finish implementation and write unit
6340 * configure.in: add --enable-tests to enable unit tests
6342 * dbus/dbus-test.c: test program to run unit tests
6343 for all files in dbus/*, initially runs a test for
6346 * dbus/dbus-internals.h: file to hold some internal utility stuff
6348 2002-11-22 Havoc Pennington <hp@redhat.com>
6350 * dbus/dbus-hash.c: copy in Tcl hash table, not yet
6351 "ported" away from Tcl
6353 * dbus/dbus-types.h: header for types such as dbus_bool_t
6355 2002-11-22 Havoc Pennington <hp@redhat.com>
6357 * dbus/dbus.h: fixups for doc warnings
6359 * Doxyfile.in (FILE_PATTERNS): we need to scan .h to pick up
6361 (QUIET): make it quiet so we can see warnings
6363 * dbus/dbus-memory.c: teach D-BUS to allocate and free memory
6365 2002-11-22 Havoc Pennington <hp@redhat.com>
6367 * Makefile.am: include "Doxyfile" target in all-local
6369 * configure.in: generate the Doxyfile
6371 * Doxyfile.in: move Doxyfile here, so we can use
6372 configure to generate a Doxyfile with the right
6375 2002-11-22 Havoc Pennington <hp@redhat.com>
6377 * dbus/dbus-message.c: move inline docs into .c file
6379 * Doxyfile (OUTPUT_DIRECTORY): move output to doc/api
6380 so all docs are under doc/
6381 (MAN_EXTENSION): generate man pages. Use extension
6382 ".3dbus" which matches ".3qt" on my system,
6383 I guess this is OK, I don't know really.
6384 (FILE_PATTERNS): look for .c files not .h, makes sense
6387 2002-11-22 Havoc Pennington <hp@pobox.com>
6389 * Makefile.am (SUBDIRS): rename subdir "server" to "bus"
6390 because any app can be a server, and any app can be a client,
6391 the bus is a special kind of server.
6393 Thu Nov 21 23:35:31 2002 Zack Rusin <zack@kde.org>
6395 * Doxyfile : adding. Still needs Makefile rules to be generated
6396 automatically (just run "doxygen" in the toplevel dir for now to
6399 * dbus/dbus-message.h : Adding sample docs (javadoc since
6400 resembles gtk-doc a little more)
6402 * dbus/dbus.h : Adding sample docs
6404 2002-11-21 Havoc Pennington <hp@redhat.com>
6406 * dbus/Makefile.am (INCLUDES): define DBUS_COMPILATION
6407 so we can allow ourselves to include files directly,
6408 instead of having to use dbus.h
6410 * dbus/dbus.h: fill in
6412 * dbus/dbus-message.h: sketch out a sample header file.
6413 Include griping if you include it directly instead of
6416 * dbus/dbus-macros.h: new file with macros for extern "C",
6417 TRUE/FALSE, NULL, etc.
6419 * doc/file-boilerplate.c: put include guards in here
6421 2002-11-21 Havoc Pennington <hp@redhat.com>
6423 * doc/file-boilerplate.c: include both AFL and GPL boilerplate.
6425 * COPYING: include the GPL as well, and license code
6426 under both AFL and GPL.
6428 2002-11-21 Havoc Pennington <hp@redhat.com>
6430 * acconfig.h: get rid of this
6432 * autogen.sh (run_configure): add --no-configure option
6434 * configure.in: remove AC_ARG_PROGRAM to make
6435 autoconf complain less. add AC_PREREQ.
6436 add AC_DEFINE third arg.
6438 2002-11-21 Anders Carlsson <andersca@codefactory.se>
6441 Fix references so we can distcheck.
6443 2002-11-21 Havoc Pennington <hp@redhat.com>
6445 * Initial module creation