1 2005-01-03 Havoc Pennington <hp@redhat.com>
3 * dbus/dbus-internals.h: I'm an idiot, _dbus_assert certainly can
6 2004-12-26 Havoc Pennington <hp@redhat.com>
8 * dbus/dbus-internals.h: add _DBUS_GNUC_NORETURN to _dbus_assert
10 2005-01-03 Havoc Pennington <hp@redhat.com>
12 * dbus/dbus-sysdeps.c (_dbus_sysdeps_test): fix using == on
15 * dbus/dbus-string.c (_dbus_string_insert_alignment): new function
17 2005-01-02 Havoc Pennington <hp@redhat.com>
19 * dbus/dbus-internals.h (_DBUS_ALIGN_OFFSET): new macro
21 2005-01-01 Havoc Pennington <hp@redhat.com>
23 * configure.in: add -Wfloat-equal
25 2005-01-01 Havoc Pennington <hp@redhat.com>
27 * dbus/dbus-sysdeps.h: add _DBUS_DOUBLES_BITWISE_EQUAL macro,
28 for a variety of reasons '==' doesn't do this.
30 2004-12-31 Havoc Pennington <hp@redhat.com>
32 * dbus/dbus-string.c (_dbus_string_equal_substrings): new function
35 2004-12-30 John (J5) Palmieri <johnp@redhat.com>
37 * python/dbus.py: s/ACTIVATION_REPLY_ACTIVE/ACTIVATION_REPLY_ACTIVATED
39 2004-12-30 John (J5) Palmieri <johnp@redhat.com>
41 * python/dbus_bindings.pyx.in: Change DBUS_ACTIVATION_REPLY_ACTIVATED
42 and DBUS_ACTIVATION_REPLY_ALREADY_ACTIVE to match the values in
43 dbus-protocol.h. Because they are defines and not enums they are not
46 2004-12-26 John (J5) Palmieri <johnp@redhat.com>
48 * python/dbus_bindings.pyx.in (bus_activate_service): Bind
49 dbus_bus_activate_service
51 * python/dbus.py (Bus.activate_service): activate a service on the
54 2004-12-24 Havoc Pennington <hp@redhat.com>
56 * test/decode-gcov.c: change to use .gcno and .gcda files, but the
57 file format has also changed and I haven't adapted to that yet
59 * Makefile.am: load .gcno files from latest gcc
61 2004-12-23 John (J5) Palmieri <johnp@redhat.com>
62 * Patch from Rob Taylor <robtaylor@fastmail.fm>
64 * python/dbus_bindings.pyx.in (bus_get_unix_user): New
67 * python/dbus.py (get_unix_user): Added binding to
68 call dbus_bindings.bus_get_unix_user
70 * python/extract.py: Modified the proto_pat regex to
73 2004-12-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
75 * dbus/make-dbus-glib-error-enum.sh: omit the function keyword for
76 better POSIX compliance.
78 2004-12-19 Havoc Pennington <hp@redhat.com>
80 * dbus/dbus-string.c (_dbus_string_insert_4_aligned)
81 (_dbus_string_insert_8_aligned): new functions
83 * dbus/dbus-string.c (_dbus_string_alloc_space): new function
85 2004-12-18 Havoc Pennington <hp@redhat.com>
87 * dbus/dbus-string.c (_dbus_string_validate_ascii): use ISASCII
90 * dbus/dbus-message.c: fix a comment, and add a still-unused
91 not-implemented function
93 * dbus/dbus-marshal.h: fix comment
95 * dbus/dbus-internals.h (_DBUS_ISASCII): new macro
97 2004-12-17 Joe Shaw <joeshaw@novell.com>
99 * mono/DBusType/Byte.cs, mono/DBusType/Int32.cs,
100 mono/DBusType/Int64.cs, mono/DBusType/UInt32.cs,
101 mono/DBusType/UInt64.cs: Use Enum.GetUnderlyingType() instead of
102 Type.UnderlyingSystemType to get the actual system type
103 underneath. This code previously depended on the broken Mono
104 behavior, which was fixed in 1.1.3.
106 2004-11-27 Havoc Pennington <hp@redhat.com>
108 * dbus/dbus-string.h (_dbus_string_get_byte): inline when asserts
110 (_dbus_string_get_const_data): inline when asserts are disabled
112 * dbus/dbus-message.c: record the _dbus_current_generation of
113 creation so we can complain if dbus_shutdown() is used improperly.
114 Do this only if checks are enabled.
116 * dbus/dbus-connection.c: ditto
118 2004-11-26 Havoc Pennington <hp@redhat.com>
120 * test/glib/test-profile.c: add with_bus mode to profile echoes
121 that go through the bus.
123 * test/glib/run-test.sh: add ability to run test-profile
125 * bus/dbus-daemon-1.1.in: fix to say that SIGHUP causes partial
128 2004-11-26 Havoc Pennington <hp@redhat.com>
130 * test/glib/test-profile.c: clean up how the fake_malloc_overhead
131 thing was implemented
133 2004-11-26 Havoc Pennington <hp@redhat.com>
135 * test/glib/test-profile.c: tweak a bit, add support for some
136 made-up minimal malloc overhead with plain sockets, since in
137 real life some sort of buffers are unavoidable thus we could
138 count them in the theoretical best case
140 2004-11-26 Havoc Pennington <hp@redhat.com>
142 * dbus/dbus-message.c (dbus_message_cache_or_finalize): fix bug
143 where I was trying to cache one too many messages
145 2004-11-26 Havoc Pennington <hp@redhat.com>
147 * dbus/dbus-message.c: reimplement message cache as an array which
148 makes the cache about twice as fast and saves maybe 1.5% overall
150 2004-11-26 Havoc Pennington <hp@redhat.com>
152 * dbus/dbus-threads.c (init_global_locks): forgot to put the
153 message cache lock here
155 2004-11-26 Havoc Pennington <hp@redhat.com>
157 * dbus/dbus-message.c (struct DBusMessage): put the locked bit and
158 the "char byte_order" next to each other to save 4 bytes
159 (dbus_message_new_empty_header): reduce preallocation, since the
160 message cache should achieve a similar effect
161 (dbus_message_cache_or_finalize, dbus_message_get_cached): add a
162 message cache that keeps a few DBusMessage around in a pool,
163 another 8% speedup or so.
165 * dbus/dbus-dataslot.c (_dbus_data_slot_list_clear): new function
167 2004-11-25 Havoc Pennington <hp@redhat.com>
169 * dbus/dbus-transport-unix.c (unix_do_iteration): if we're going
170 to write, without reading or blocking, try it before the poll()
171 and skip the poll() if nothing remains to write. This is about a
172 3% speedup in the echo client/server
174 2004-11-25 Havoc Pennington <hp@redhat.com>
176 The primary change here is to always write() once before adding
177 the write watch, which gives us about a 10% performance increase.
179 * dbus/dbus-transport-unix.c: a number of modifications to cope
180 with removing messages_pending
181 (check_write_watch): properly handle
182 DBUS_AUTH_STATE_WAITING_FOR_MEMORY; adapt to removal of
183 messages_pending stuff
184 (check_read_watch): properly handle WAITING_FOR_MEMORY and
186 (unix_handle_watch): after writing, see if the write watch can be
188 (unix_do_iteration): assert that write_watch/read_watch are
189 non-NULL rather than testing that they aren't, since they
190 aren't allowed to be NULL. check_write_watch() at the end so
191 we add the watch if we did not finish writing (e.g. got EAGAIN)
193 * dbus/dbus-transport-protected.h: remove messages_pending call,
194 since it resulted in too much inefficient watch adding/removing;
195 instead we now require that the transport user does an iteration
196 after queueing outgoing messages, and after trying the first
197 write() we add a write watch if we got EAGAIN or exceeded our
198 max bytes to write per iteration setting
200 * dbus/dbus-string.c (_dbus_string_validate_signature): add this
203 * dbus/dbus-server-unix.c (unix_finalize): the socket name was
204 freed and then accessed, valgrind flagged this bug, fix it
206 * dbus/dbus-message.c: fix several bugs where HEADER_FIELD_LAST was taken
207 as the last valid field plus 1, where really it is equal to the
208 last valid field. Corrects some message corruption issues.
210 * dbus/dbus-mainloop.c: verbosity changes
212 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): handle OOM
213 instead of aborting in one of the test codepaths
215 * dbus/dbus-internals.c (_dbus_verbose_real): fix a bug that
216 caused not printing the pid ever again if a verbose was missing
217 the newline at the end
218 (_dbus_header_field_to_string): add HEADER_FIELD_SIGNATURE
220 * dbus/dbus-connection.c: verbosity changes;
221 (dbus_connection_has_messages_to_send): new function
222 (_dbus_connection_message_sent): no longer call transport->messages_pending
223 (_dbus_connection_send_preallocated_unlocked): do one iteration to
224 try to write() immediately, so we can avoid the write watch. This
225 is the core purpose of this patchset
226 (_dbus_connection_get_dispatch_status_unlocked): if disconnected,
227 dump the outgoing message queue, so nobody will get confused
228 trying to send them or thinking stuff is pending to be sent
230 * bus/test.c: verbosity changes
232 * bus/driver.c: verbosity/assertion changes
234 * bus/dispatch.c: a bunch of little tweaks to get it working again
235 because this patchset changes when/where you need to block.
237 2004-11-23 Havoc Pennington <hp@redhat.com>
239 * test/glib/test-profile.c: modify to accept a plain_sockets
240 argument in which case it will bench plain sockets instead of
241 libdbus, for comparison purposes.
243 2004-11-22 Havoc Pennington <hp@redhat.com>
245 * test/glib/test-profile.c (N_CLIENT_THREADS): run multiple
246 threads for more time, so sysprof can get a grip on it.
248 * dbus/dbus-string.c (_dbus_string_validate_utf8): remove
251 2004-11-13 Havoc Pennington <hp@redhat.com>
253 * test/glib/test-profile.c: fix this thing up a bit
255 * dbus/dbus-message.c (dbus_message_new_empty_header): increase
256 preallocation sizes by a fair bit; not sure if this will be an
257 overall performance win or not, but it does reduce reallocs.
259 * dbus/dbus-string.c (set_length, reallocate_for_length): ignore
260 the test hack that forced constant realloc if asserts are
261 disabled, so we can profile sanely. Sprinkle in some
262 _DBUS_UNLIKELY() which are probably pointless, but before I
263 noticed the real performance problem I put them in.
264 (_dbus_string_validate_utf8): micro-optimize this thing a little
265 bit, though callgrind says it didn't help; then special-case
266 ascii, which did help a lot; then be sure we detect nul bytes as
267 invalid, which is a bugfix.
268 (align_length_then_lengthen): add some more _DBUS_UNLIKELY
269 superstition; use memset to nul the padding instead of a manual
271 (_dbus_string_get_length): inline this as a
272 macro; it showed up in the profile because it's used for loop
275 2004-11-10 Colin Walters <walters@verbum.org>
277 * dbus/dbus-spawn.c (check_babysit_events): Handle EINTR,
280 2004-11-09 Colin Walters <walters@verbum.org>
282 * dbus/dbus-string.c (_dbus_string_get_length): New
283 function, writes DBusString to C buffer.
285 * dbus/dbus-string.h: Prototype it.
287 * dbus/dbus-message.c (dbus_message_type_to_string): New
288 function, converts message type into C string.
290 * dbus/dbus-message.h: Prototype it.
292 * bus/selinux.c (bus_selinux_check): Take source pid,
293 target pid, and audit data. Pass audit data to
295 (log_audit_callback): New function, appends extra
297 (bus_selinux_allows_acquire_service): Also take
298 service name, add it to audit data.
299 (bus_selinux_allows_send): Also take message
300 type, interface, method member, error name,
301 and destination, and add them to audit data.
302 (log_cb): Initialize func_audit.
304 * bus/selinux.h (bus_selinux_allows_acquire_service)
305 (bus_selinux_allows_send): Update prototypes
307 * bus/services.c (bus_registry_acquire_service): Pass
308 service name to bus_selinux_allows_acquire_service.
310 * bus/bus.c (bus_context_check_security_policy): Pass
311 additional audit data. Move assignment of dest
314 2004-11-07 Colin Walters <walters@verbum.org>
316 * dbus/dbus-transport-unix.c (do_authentication): Always
317 initialize auth_completed.
319 2004-11-07 Colin Walters <walters@verbum.org>
321 * bus/bus.c (load_config): Break into three
322 separate functions: process_config_first_time_only,
323 process_config_every_time, and process_config_postinit.
324 (process_config_every_time): Move call of
325 bus_registry_set_service_context_table into
326 process_config_postinit.
327 (process_config_postinit): New function, does
328 any processing that needs to happen late
329 in initialization (and also on reload).
330 (bus_context_new): Instead of calling load_config,
331 open config parser here and call process_config_first_time_only
332 and process_config_every_time directly. Later, after
333 we have forked but before changing UID,
334 invoke bus_selinux_full_init, and then call
335 process_config_postinit.
336 (bus_context_reload_config): As in bus_context_new,
337 load parse file inside here, and call process_config_every_time
338 and process_config_postinit.
340 * bus/services.h, bus/services.c
341 (bus_registry_set_service_context_table): Rename
342 from bus_registry_set_sid_table. Take string hash from config
343 parser, and convert them here into SIDs.
345 * bus/config-parser.c (struct BusConfigParser): Have
346 config parser only store a mapping of service->context
348 (merge_service_context_hash): New function.
349 (merge_included): Merge context string hashes instead
350 of using bus_selinux_id_table_union.
351 (bus_config_parser_new): Don't use bus_selinux_id_table_new;
352 simply create a new string hash.
353 (bus_config_parser_unref): Unref it.
354 (start_selinux_child): Simply insert strings into hash,
355 don't call bus_selinux_id_table_copy_over.
357 * bus/selinux.h, bus/selinux.c (bus_selinux_id_table_union)
358 (bus_selinux_id_table_copy_over): Delete.
360 2004-11-03 Colin Walters <walters@verbum.org>
362 * bus/selinux.c (bus_selinux_pre_init): Kill some unused
365 2004-11-03 Colin Walters <walters@verbum.org>
367 * bus/test-main.c (test_pre_hook): Fix test logic,
368 thanks Joerg Barfurth <Joerg.Barfurth@Sun.COM>.
370 2004-11-02 Colin Walters <walters@redhat.com>
372 * bus/selinux.c (bus_selinux_init): Split into two functions,
373 bus_selinux_pre_init and bus_selinux_post_init.
374 (bus_selinux_pre_init): Just determine whether SELinux is
376 (bus_selinux_post_init): Do everything else.
378 * bus/main.c (main): Call bus_selinux_pre_init before parsing
379 config file, and bus_selinux_post_init after. This ensures that
380 we don't lose the policyreload notification thread that
381 bus_selinux_init created before forking previously.
383 * bus/test-main.c (test_pre_hook): Update for split.
385 2004-10-31 Owen Fraser-Green <owen@discobabe.net>
387 Patch from Johan Fischer <linux@fischaz.com>
389 * mono/doc/Makefile.am (install-data-local): Added directory
392 2004-10-29 Colin Walters <walters@redhat.com>
394 * dbus/dbus-sysdeps.h (_dbus_become_daemon): Also take
395 parameter for fd to write pid to.
397 * dbus/dbus-sysdeps.c (_dbus_become_daemon): Implement it.
399 * bus/bus.c (bus_context_new): Pass print_pid_fd
400 to _dbus_become_daemon (bug #1720)
402 2004-10-29 Colin Walters <walters@redhat.com>
404 Patch from Ed Catmur <ed@catmur.co.uk>
406 * mono/doc/Makefile.am (install-data-local): Handle
409 2004-10-29 Colin Walters <walters@redhat.com>
411 * bus/.cvsignore, qt/.cvsignore: Update.
413 2004-10-29 Colin Walters <walters@redhat.com>
415 Patch from Kristof Vansant <de_lupus@pandora.be>
417 * configure.in: Detect Slackware.
418 * bus/Makefile.am (SCRIPT_IN_FILES): Add rc.messagebus.in.
419 * bus/rc.messagebus.in: New file.
421 2004-10-29 Colin Walters <walters@redhat.com>
423 * tools/dbus-monitor.c (filter_func): Return
424 DBUS_HANDLER_RESULT_HANDLED in filter function
426 http://freedesktop.org/pipermail/dbus/2004-August/001433.html
428 2004-10-29 Colin Walters <walters@redhat.com>
430 Patch from Matthew Rickard <mjricka@epoch.ncsc.mil>
432 * bus/services.c (bus_registry_acquire_service):
433 Correctly retrieve service name from DBusString
436 2004-10-29 Colin Walters <walters@redhat.com>
438 * dbus/dbus-glib.h: Update documentation to not
439 refer to internal APIs.
441 2004-10-27 Joe Shaw <joeshaw@novell.com>
443 * mono/Arguments.cs (GetDBusTypeConstructor):
444 type.UnderlyingSystemType will return "System.Byte" if you do it
445 on "byte[]", which is not what we want. So check the type.IsArray
446 property and use System.Array instead.
448 2004-10-25 John (J5) Palmieri <johnp@redhat.com>
450 * dbus/dbus-sysdeps.c (fill_user_info): On errors do not free
451 the DBusUserInfo structure since this is passed into the function.
452 This would cause a double free when the function that allocated
453 the structure would try to free it when an error occured.
455 * (bus/session.conf.in, bus/Makefile.am, dbus/configure.in):
456 use /usr/share/dbus-1/services instead of /usr/lib/dbus-1.0/services
457 for service activation to avoid 32bit/64bit parallel install issues
459 2004-10-21 Colin Walters <walters@verbum.org>
461 * AUTHORS: Fix my email address, the @gnu.org one
462 has been bouncing for some time. Also add J5.
464 2004-10-21 Colin Walters <walters@verbum.org>
466 * dbus/dbus-transport-unix.c (do_authentication): Return
467 authentication status to callers.
468 (unix_handle_watch): If we completed authentication this round,
469 don't do another read. Instead wait until the next iteration,
470 after we've read any pending data in the auth buffer.
471 (unix_do_iteration): Ditto.
472 (unix_handle_watch): Updated for new do_authentication prototype.
474 2004-10-18 Colin Walters <walters@verbum.org>
476 * bus/selinux.c (bus_selinux_enabled): Handle
477 --disable-selinux case.
479 2004-10-18 Colin Walters <walters@verbum.org>
481 * bus/selinux.h: Add bus_selinux_enabled.
483 * bus/selinux.c (bus_selinux_enabled): Implement it.
485 * bus/config-parser.c (struct include): Add
486 if_selinux_enabled member.
487 (start_busconfig_child): Parse if_selinux_enabled
488 attribute for include.
489 (bus_config_parser_content): Handle it.
491 * bus/session.conf.in, bus/system.conf.in: Add
492 inclusion of context mapping to default config files;
493 conditional on SELinux being enabled.
495 * doc/busconfig.dtd: Add to if_selinux_enabled to default DTD.
497 * test/data/invalid-config-files/badselinux-1.conf,
498 test/data/invalid-config-files/badselinux-2.conf:
499 Test files for bad syntax.
501 2004-10-17 Colin Walters <walters@verbum.org>
503 * dbus/dbus-memory.c (_dbus_initialize_malloc_debug, check_guards)
504 (dbus_malloc, dbus_malloc0, dbus_realloc): Fix up printf
505 format specifier mismatches.
507 2004-10-07 Olivier Andrieu <oliv__a@users.sourceforge.net>
509 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): fix an incorrect
512 * glib/dbus-dbus-gmain.c (dbus_g_bus_get): do not mangle NULL
513 pointer (bug #1540, Leonardo Boiko).
515 2004-09-28 Jon Trowbridge <trow@ximian.com>
517 * mono/BusDriver.cs: Changed BusDriver struct to remove
518 the ServiceCreated and ServiceDeleted events and replace them
519 with the new ServiceOwnerChanged event.
521 * mono/example/BusListener.cs: Added a new example program,
522 which listens for and reports any ServiceOwnerChanged events
525 * mono/example/Makefile.am (DESTDIR): Build changes for the
526 new BusListener.cs example.
528 2004-09-27 Olivier Andrieu <oliv__a@users.sourceforge.net>
530 * bus/signals.c (bus_match_rule_parse): validate the components of
531 match rules (bug #1439).
533 * dbus/dbus-bus.c (dbus_bus_add_match): add a missing OOM test.
535 2004-09-24 Olivier Andrieu <oliv__a@users.sourceforge.net>
537 * doc/dbus-specification.xml: document ServiceOwnerChanged
540 * bus/driver.c, bus/driver.h, bus/services.c: Use
541 ServiceOwnerChanged signal instead of ServiceCreated and
544 * bus/dispatch.c: update testcase for the new signal.
546 2004-09-20 Jon Trowbridge <trow@ximian.com>
548 Patch from Nat Friedman <nat@novell.com>
550 * mono/Makefile.am: A number of small build fixes to allow "make
551 distcheck" to succeed.
553 * mono/example/Makefile.am: "make distcheck" fixes.
555 * mono/AssemblyInfo.cs.in: When signing the assembly, look for the
558 * test/Makefile.am: "make distcheck" fixes.
560 2004-09-17 Olivier Andrieu <oliv__a@users.sourceforge.net>
562 * dbus/dbus-sysdeps.c (_dbus_user_at_console): fix memleak in OOM.
564 * doc/busconfig.dtd: update the DTD for the at_console attribute.
566 * bus/driver.c (bus_driver_handle_hello): correctly handle Hello
567 messages after the first one (bug #1389).
569 * bus/dispatch.c (check_double_hello_message): add a test case for
570 the double hello message bug.
571 (check_existent_service_activation): fix check of spawning error.
573 2004-09-16 David Zeuthen <david@fubar.dk>
575 * python/dbus_bindings.pyx.in: Add support for int64 and uint64
577 2004-09-12 David Zeuthen <david@fubar.dk>
579 Patch from Kay Sievers <kay.sievers@vrfy.org>
581 * bus/bus.c (bus_context_new):
585 Add commandline option --nofork to override configuration file
588 2004-09-09 Olivier Andrieu <oliv__a@users.sourceforge.net>
590 * dbus/dbus-*.h: remove the ; after DBUS_(BEGIN|END)_DECLS. Some C
591 compilers don't like it (bug #974).
593 2004-09-04 Harald Fernengel <harry@kdevelop.org>
595 * qt/connection.*: Applied patch by Jérôme Lodewyck
596 <lodewyck@clipper.ens.fr> to integrate an existing
597 connection into the Qt eventloop
599 2004-08-30 Jon Trowbridge <trow@ximian.com>
601 * mono/BusDriver.cs: Added. This is a class for interacting with
602 the org.freedesktop.DBus service.
604 * mono/Message.cs: Added a mechanism to expose the message that is
605 currently being dispatched via the static Message.Current
606 property. Added Message.Sender and Message.Destination
609 * mono/Handler.cs: Expose the dispatched message via
610 Message.Current when handling method calls.
612 * mono/Service.cs: Expose the dispatched message via
613 Message.Current when handling signal emissions.
615 * mono/Connection.cs: Bind dbus_bus_get_base_service via the
616 Connection.BaseService property.
618 2004-08-28 Havoc Pennington <hp@redhat.com>
620 * dbus/dbus-userdb.c (_dbus_is_console_user): remove unused variable
622 More fixes from Steve Grubb
624 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): fix fd leak
625 (_dbus_listen_tcp_socket): fix fd leak
627 * dbus/dbus-spawn.c (read_pid, read_ints): move the "again:" for
628 EINTR to a bit lower in the code
630 2004-08-26 Jon Trowbridge <trow@ximian.com>
632 * bus/driver.c (bus_driver_handle_service_exists): Respond with
633 TRUE if we are inquiring about the existence of the built-in
634 org.freedesktop.DBus service.
636 2004-08-25 John Palmieri <johnp@redhat.com>
637 * bus/config-parser.c:
638 (struct PolicyType): Add POLICY_CONSOLE
639 (struct Element.d.policy): s/gid_or_uid/gid_uid_or_at_console
640 (start_busconfig_child): Sets up console element when
641 <policy at_console=""> is encountered in a policy file
642 (append_rule_from_element): Convert console elements to console
646 (bus_policy_create_client_policy): Add console rules to the client
647 policy based on if the client is at the console
648 (bus_policy_append_console_rule): New function for adding a
649 console rule to a policy
650 (bus_policy_merge): Handle console rule merging
652 * dbus/dbus-sysdeps.h: Added the DBUS_CONSOLE_DIR constant
653 where we check for console user files
655 * dbus/dbus-sysdeps.c:
656 (_dbus_file_exists): New function which checks if the given
658 (_dbus_user_at_console): New function which does the system
659 specific process of checking if the user is at the console
661 * dbus/dbus-userdb.c:
662 (_dbus_is_console_user): New function converts a UID to user name
663 and then calls the system specific _dbus_user_at_console to
664 see if the user is at the console and therefor a console user
666 2004-08-25 Olivier Andrieu <oliv__a@users.sourceforge.net>
668 * bus/config-parser.c (set_limit):
669 * bus/dbus-daemon-1.1.in:
670 * test/data/valid-config-files/many-rules.conf: set the
671 max_match_rules_per_connection limt from the config file.
673 * doc/busconfig.dtd: update the DTD.
675 * bus/driver.c: remove some unused variables.
677 2004-08-24 Mikael Hallendal <micke@imendio.com>
679 * dbus/dbus-glib-lowlevel.h: Removed dbus_bus_get_with_g_main since
680 it's been replaced by dbus_g_bus_get
682 2004-08-23 Colin Walters <walters@redhat.com>
684 Updated SELinux support from Matthew Rickard <mjricka@epoch.ncsc.mil>
686 * bus/selinux.h: Prototype bus_selinux_get_policy_root.
688 * bus/selinux.c: Create a thread for policy reload notification.
689 (bus_selinux_get_policy_root): Implement.
691 * bus/config-parser.c (start_busconfig_child)
692 (bus_config_parser_content): Support SELinux-root relative
695 * configure.in <HAVE_SELINUX>: Add -lpthread.
697 * bus/test-main.c (test_pre_hook, test_post_hook): New.
698 (test_post_hook): Move memory checking into here.
699 (test_pre_hook, test_post_hook): Move SELinux checks in
700 here, but conditional on a DBUS_TEST_SELINUX environment
701 variable. Unfortunately we can't run the SELinux checks
702 as a normal user, since they won't have any permissions
703 for /selinux. So this will have to be tested manually
704 for now, until we have virtualization for most of
707 2004-08-23 Havoc Pennington <hp@redhat.com>
709 * dbus/dbus-sysdeps.c (_dbus_change_identity): add setgroups() to
710 drop supplementary groups, suggested by Steve Grubb
712 2004-08-20 Colin Walters <walters@redhat.com>
714 * bus/config-parser.c (start_busconfig_child): Remove some unused
717 * bus/selinux.c (bus_selinux_id_table_insert): Avoid compiler
720 2004-08-17 Joe Shaw <joeshaw@novell.com>
722 * configure.in: If --enable-mono is passed in, if we can't find
725 * mono/Makefile.am: Use /gacutil to install assemblies into the
728 2004-08-12 Havoc Pennington <hp@redhat.com>
730 * NEWS: update for 0.22
732 * configure.in: release 0.22
734 2004-08-11 Colin Walters <walters@redhat.com>
736 * tools/dbus-send.c (main, usage): Add --reply-timeout
739 2004-08-10 Olivier Andrieu <oliv__a@users.sourceforge.net>
741 * bus/bus.c (process_config_first_time_only): get rid of an unused
742 DBusError that was causing a memoy leak (bug #989).
744 * dbus/dbus-keyring.c, dbus/dbus-message.c:
745 fix compilation on Solaris/Forte C (bug #974)
747 * bus/main.c (main): plug two minuscule memleaks.
749 2004-08-10 Havoc Pennington <hp@redhat.com>
751 * doc/dbus-tutorial.xml: add some more info on GLib bindings
753 2004-08-09 Havoc Pennington <hp@redhat.com>
755 * COPYING: switch to Academic Free License version 2.1 instead of
756 2.0, to resolve complaints about patent termination clause.
758 2004-07-31 John (J5) Palmieri <johnp@redhat.com>
760 * README: added documentation for the --enable-python
763 2004-07-31 Olivier Andrieu <oliv__a@users.sourceforge.net>
765 * bus/config-parser.c (bus_config_parser_new): fix an invalid
766 _unref in the SELinux support.
768 * doc/busconfig.dtd: update DTD for SELinux support.
770 * bus/config-loader-libxml.c: fix error handler and parser
771 initialisation/cleanup. OOM test now works with libxml2 HEAD.
773 * configure.in: remove the warning about libxml2.
775 * dbus/dbus-bus.c: silence doxygen warning.
777 2004-07-31 Colin Walters <walters@redhat.com>
779 * configure.in: Move #error in SELinux check to its own line.
781 2004-07-31 Olivier Andrieu <oliv__a@users.sourceforge.net>
783 * dbus/dbus-internals.h (_DBUS_SET_OOM):
784 * bus/utils.h (BUS_SET_OOM): use dbus_error_set_const instead of
787 * bus/dispatch.c (check_send_exit_to_service): fix the test case,
788 broken by the change in the _SET_OOM macros.
790 2004-07-31 Colin Walters <walters@redhat.com>
792 * bus/selinux.c <HAVE_SELINUX>: Include utils.h to get
795 2004-07-31 Colin Walters <walters@redhat.com>
797 * configure.in: Use AC_TRY_COMPILE instead of AC_EGREP_HEADER
798 to correctly detect DBUS__ACQUIRE_SVC. Also add an
801 2004-07-24 Havoc Pennington <hp@redhat.com>
803 SELinux support from Matthew Rickard <mjricka@epoch.ncsc.mil>
805 * bus/selinux.c, bus/selinux.h: new file encapsulating selinux
808 * configure.in: add --enable-selinux
810 * bus/policy.c (bus_policy_merge): add FIXME to a comment
812 * bus/main.c (main): initialize and shut down selinux
814 * bus/connection.c: store SELinux ID on each connection, to avoid
815 repeated getting of the string context and converting it into
818 * bus/bus.c (bus_context_get_policy): new accessor, though it
820 (bus_context_check_security_policy): check whether the security
821 context of sender connection can send to the security context of
824 * bus/config-parser.c: add parsing for <selinux> and <associate>
826 * dbus/dbus-transport.c (_dbus_transport_get_unix_fd): to
827 implement dbus_connection_get_unix_fd()
829 * dbus/dbus-connection.c (dbus_connection_get_unix_fd): new
830 function, used by the selinux stuff
832 2004-07-29 Olivier Andrieu <oliv__a@users.sourceforge.net>
834 * bus/config-loader-libxml.c: complete the implementation of
835 libxml backend for config file loader. Doesn't work with full OOM
838 * configure.in: change error when selecting libxml into a warning.
840 * test/data/invalid-config-files: add two non-well-formed XML
843 * glib/Makefile.am: libdbus_gtool always uses expat, not libxml.
845 * dbus/dbus-transport-unix.c (unix_handle_watch): do not
846 disconnect in case of DBUS_WATCH_HANGUP, several do_reading() may
847 be necessary to read all the buffer. (bug #894)
849 * bus/activation.c (bus_activation_activate_service): fix a
850 potential assertion failure (bug #896). Small optimization in the
851 case of auto-activation messages.
853 * dbus/dbus-message.c (verify_test_message, _dbus_message_test):
854 add test case for byte-through-vararg bug (#901). patch by Kimmo
857 2004-07-28 Anders Carlsson <andersca@gnome.org>
860 * python/dbus_bindings.pyx.in:
861 Add dbus.init_gthreads (), allow emit_signal to pass
862 arguments to the signal.
864 2004-07-24 Havoc Pennington <hp@redhat.com>
866 * AUTHORS: add some people, not really comprehensively, let me
869 2004-07-24 Havoc Pennington <hp@redhat.com>
871 * Makefile.am (DIST_SUBDIRS): add DIST_SUBDIRS, problem solved by
874 * test/Makefile.am (DIST_SUBDIRS): here also
876 2004-07-22 Olivier Andrieu <oliv__a@users.sourceforge.net>
878 * dbus/dbus-sysdeps.c (fill_user_info): fix inexistent label name,
879 breaking build on Solaris, reported by Farhad Saberi on the ML.
881 * dbus/dbus-message.c (dbus_message_append_args_valist): fix the
882 va_arg invocation to account for integer promotion in the case of
883 DBUS_TYPE_BYTE (unsigned char is promoted to int). (bug #901)
885 * bus/services.c (bus_service_remove_owner): fix bug #902, use
886 _dbus_list_get_first_link, not _dbus_list_get_first.
888 * dbus/dbus-bus.c (dbus_bus_service_exists): plug a memory leak.
890 * dbus/dbus-object-tree.c (free_subtree_recurse): always null
891 handler functions so that the asserts in _dbus_object_subtree_unref
894 * dbus/dbus-transport-unix.c (do_reading):
895 _dbus_transport_queue_messages return value is of type
896 dbus_bool_t, not DBusDispatchStatus.
898 2004-07-19 David Zeuthen <david@fubar.dk>
900 * dbus/dbus-protocol.h: Add DBUS_ERROR_UNIX_PROCESS_ID_UNKNOWN
903 (check_get_connection_unix_user): Debug says GetProperty; but the
904 method is called GetConnectionUnixUser
905 (check_get_connection_unix_process_id): New function
906 (bus_dispatch_test): Actually call check_get_connection_unix_user();
907 also call check_get_connection_unix_process_id()
910 (bus_driver_handle_get_connection_unix_process_id): New function,
911 handles GetConnectionUnixProcessID on the org.freedesktop.DBus
915 (handle_server_data_external_mech): Set pid from the credentials
916 obtained from the socket
918 * dbus/dbus-connection.c:
919 (dbus_connection_get_unix_process_id): New function
921 * dbus/dbus-connection.h:
922 Add prototype for dbus_connection_get_unix_process_id
924 * dbus/dbus-transport.c:
925 (_dbus_transport_get_unix_process_id): New function
927 * dbus/dbus-transport.h:
928 Add prototype for _dbus_transport_get_unix_process_id
930 2004-07-19 Olivier Andrieu <oliv__a@users.sourceforge.net>
932 * dbus/dbus-message.c: Message counter fix, patch by Christian
933 Hammond <chipx86@gnupdate.org>
935 2004-07-18 Seth Nickell <seth@gnome.org>
938 * python/dbus_bindings.pyx.in:
939 * python/tests/test-client.py:
941 Add dbus.ByteArray and dbus_bindings.ByteArray
942 types so that byte streams can be passed back.
944 Give jdahlin the heaps of credit that are so
947 2004-07-12 Seth Nickell <seth@gnome.org>
951 Add message argument to the default object_method_handler
954 * python/dbus_bindings.pyx.in:
956 Automatically return NIL when passed an empty list
957 (we can't pass back a list since lists are typed
958 and we don't have any idea what type the the client
959 intended the list to be... :-( )
961 2004-07-10 Seth Nickell <seth@gnome.org>
963 * python/examples/Makefile.am:
965 Fix distcheck breakage caused by new examples.
967 2004-07-10 Seth Nickell <seth@gnome.org>
971 Add "message" argument to service-side dbus.Object
972 methods. This will break existing services written
973 using the python bindings, but will allow extraction
974 of all the message information (e.g. who its from).
976 Add improved "object oriented" signal handling/emission.
978 * python/examples/example-service.py:
982 * python/examples/example-signal-emitter.py:
983 * python/examples/example-signal-recipient.py:
985 Two new examples that show how to emit and receive
986 signals using the new APIs.
988 * python/examples/example-signals.py:
989 * python/examples/gconf-proxy-service.py:
990 * python/examples/gconf-proxy-service2.py:
992 Add "message" argument to service methods.
994 2004-06-28 Kay Sievers <kay.sievers@vrfy.org>
996 * bus/driver.c (bus_driver_handle_get_connection_unix_user)
997 * dbus/bus.c (dbus_bus_get_unix_user)
998 * doc/dbus-specification.xml: implement GetConnectionUnixUser
999 method of org.freedesktop.DBus interface.
1001 * bus/dispatch.c: test case
1003 2004-06-23 John (J5) Palmieri <johnp@redhat.com>
1005 * python/Makefile.am: switched include directory from glib/ to dbus/
1006 since dbus-glib.h moved
1008 2004-06-22 Olivier Andrieu <oliv__a@users.sourceforge.net>
1010 * configure.in: prevent building the gcj stuff and libxml loader
1011 since they are broken.
1013 2004-06-20 Havoc Pennington <hp@redhat.com>
1015 * dbus/dbus-glib-error-enum.h: autogenerate the GError enum
1016 codes from the dbus error names
1018 * glib/dbus-glib.h: move to subdir dbus/ since it's included
1019 as dbus/dbus-glib.h and that breakage is now visible due to
1020 including dbus/dbus-glib.h in dbus-glib-lowlevel.h
1022 * glib/dbus-glib.h: s/gproxy/g_proxy/
1024 * dbus/dbus-shared.h: new header to hold stuff shared with
1027 * dbus/dbus-protocol.h (DBUS_ERROR_*): move errors here rather
1030 * glib/dbus-glib.h (dbus_set_g_error): move to
1031 dbus-glib-lowlevel.h
1033 * glib/dbus-glib.h: remove dbus/dbus.h from here; change a bunch
1034 of stuff to enable this
1036 * dbus/dbus-glib-lowlevel.h: put dbus/dbus.h here
1038 * a bunch of other changes with the same basic "separate glib
1039 bindings from dbus.h" theme
1041 2004-06-10 Owen Fraser-Green <owen@discobabe.net>
1043 * dbus-sharp.pc.in: Removed glib-sharp inclusion in Libs.
1045 * python/examples/Makefile.am: Fixed typo in EXTRA_DIST.
1047 2004-06-09 Olivier Andrieu <oliv__a@users.sourceforge.net>
1049 * bus/driver.c, dbus/dbus-bus.c: use BOOLEAN instead of UINT32 for
1050 the reply value of the ServiceExists message.
1052 2004-06-07 John (J5) Palmieri <johnp@redhat.com>
1054 * python/dbus_bindings.pyx.in: No longer need to parse path
1055 elements and pass them as arrays of strings. The C API now
1056 accepts plain path strings.
1057 (_build_parsed_path): removed
1059 2004-06-07 Havoc Pennington <hp@redhat.com>
1061 * doc/TODO: remove auto-activation item since it's done; sort
1062 items by importance/milestone
1064 2004-06-07 Havoc Pennington <hp@redhat.com>
1066 * dbus/dbus-message-builder.c (_dbus_message_data_load): append
1067 random signature when using REQUIRED_FIELDS (this hack won't work
1070 * dbus/dbus-message.c: change the signature to be a header field,
1071 instead of message->signature special-case string. Incremental
1072 step forward. Then we can fix up code to send the signature in the
1073 message, then fix up code to validate said signature, then fix up
1074 code to not put the typecodes inline, etc.
1075 (load_one_message): don't make up the signature after the fact
1076 (decode_header_data): require signature field for the known
1079 * dbus/dbus-marshal.c (_dbus_marshal_string_len): new
1081 * dbus/dbus-protocol.h: add DBUS_HEADER_FIELD_SIGNATURE
1083 2004-06-07 Owen Fraser-Green <owen@discobabe.net>
1085 * mono/DBusType/ObjectPath.cs: Renamed PathName argument to Path
1087 * mono/Handler.cs: Updated to follow new path argument for
1088 (un-)registering objects.
1090 * mono/example/Makefile.am:
1092 * configure.in: Bumped required version for mono and use new -pkg
1095 2004-06-05 Olivier Andrieu <oliv__a@users.sourceforge.net>
1097 * dbus/dbus-connection.h, dbus/dbus-connection.c: have object path
1098 registration functions take the path argument as char* instead of
1101 * dbus/dbus-marshal.h, dbus/dbus-marshal.c (_dbus_decompose_path):
1102 split off the path decompostion part of
1103 _dbus_demarshal_object_path. Some misc. fixes to silence compiler
1106 * glib/dbus-gobject.c, test/test-service.c: update accordingly.
1108 2004-06-02 Kristian Høgsberg <krh@redhat.com>
1110 * dbus/dbus-auth.c: Rewrite auth protocol handling to use a state
1111 machine approach. A state is implemented as a function that
1112 handles incoming events as specified for that state.
1114 * doc/dbus-specification.xml: Update auth protocol state machine
1115 specification to match implementation. Remove some leftover
1118 2004-06-02 Kristian Høgsberg <krh@redhat.com>
1120 * glib/dbus-gproxy.c, glib/dbus-gmain.c, dbus/dbus-string.c,
1121 dbus/dbus-object-tree.c, dbus/dbus-message.c: add comments to
1124 * Doxyfile.in: remove deprecated options.
1126 * dbus/dbus-message-handler.c, dbus/dbus-message-handler.h,
1127 glib/test-thread.h, glib/test-thread-client.c,
1128 glib/test-thread-server.c, glib/test-profile.c,
1129 glib/test-dbus-glib.c: remove these unused files.
1131 2004-06-01 Olivier Andrieu <oliv__a@users.sourceforge.net>
1133 * dbus/dbus-object-tree.c
1134 (_dbus_object_tree_dispatch_and_unlock): fix dispatch for
1135 non-fallback handlers (bug #684).
1136 (_dbus_object_subtree_new): initialize invoke_as_fallback field.
1137 (find_subtree_recurse): report wether the returned subtree is an
1138 exact match or a "fallback" match higher up in the tree.
1139 (object_tree_test_iteration): update test case.
1141 2004-06-01 Seth Nickell <seth@gnome.org>
1143 * python/dbus_bindings.pyx.in:
1144 * python/tests/test-client.py:
1146 Round off basic type support. Add dicts (yay!), and
1147 remaining array types.
1149 Make MessageIter more general so it works for dicts too.
1151 Mark all loop variables as C integers.
1153 2004-05-31 Havoc Pennington <hp@redhat.com>
1155 * glib/dbus-gidl.c (method_info_add_arg): keep args sorted with
1158 * glib/dbus-gobject.c (dbus_type_to_string): move to dbus-gutils.c
1160 * glib/dbus-glib-tool.c (main): set up to have a --self-test
1161 option that runs the tests, and start filling in some code
1162 including for starters just dumping the interfaces to stdout
1164 * glib/Makefile.am (INCLUDES): define DBUS_LOCALEDIR
1166 * test/data/valid-introspection-files/lots-of-types.xml: test of
1167 an example introspection file
1169 * glib/dbus-gparser.c (parser_check_doctype): doctype should be
1172 2004-05-31 Seth Nickell <seth@gnome.org>
1174 * python/dbus_bindings.pyx.in:
1175 * python/tests/test-client.py:
1180 Fix string array memory trashing bug... oops...
1182 2004-05-30 Seth Nickell <seth@gnome.org>
1186 Add a nicer-but-less-flexible alternate API for handling
1187 calls to virtual objects in dbus.ObjectTree.
1189 Screw up the argument order to the dbus.Object constructor
1190 for consistency with dbus.ObjectTree (and to make dbus_methods
1191 optional for future extension)
1193 * python/examples/Makefile.am:
1194 * python/examples/gconf-proxy-service.py:
1195 * python/examples/gconf-proxy-service2.py:
1197 Alternate implementation of gconf-proxy-service using the
1198 nicer dbus.ObjectTree API.
1200 * python/examples/example-service.py:
1201 * python/tests/test-server.py
1203 Reverse the argument order to deal with dbus.Object constructor
1206 2004-05-30 Seth Nickell <seth@gnome.org>
1208 * python/examples/example-client.py:
1209 * python/examples/example-service.py:
1211 Take it back. Lists seem to work but they're broken
1212 in the test suite. Make the base examples use
1215 2004-05-30 Seth Nickell <seth@gnome.org>
1217 * python/dbus_bindings.pyx.in:
1218 * python/tests/test-client.py:
1220 Add some more tests and fix errors that crop up.
1221 Unfortunately, currently it seems like marshalling
1222 and unmarshalling of lists is completely broken :-(
1224 2004-05-30 Seth Nickell <seth@gnome.org>
1226 * python/dbus_bindings.pyx.in:
1228 Add support for ObjectPath type.
1232 Refactor message handling code to a common function.
1234 * python/tests/test-client.py:
1235 * python/tests/test-server.py:
1237 Add tests that check to make sure values of all types
1238 can be echoed from a service w/o mangling.
1240 2004-05-29 Seth Nickell <seth@gnome.org>
1244 Add ObjectTree class which allows implementation
1245 of trees of "virtual" objects. Basically the python
1246 wrapper for "register_fallback".
1248 * python/examples/Makefile.am
1249 * python/examples/gconf-proxy-client.py:
1250 * python/examples/gconf-proxy-service.py:
1252 Implement a simple GConf proxy service that supports
1253 get/set on string and int GConf keys using the ObjectTree.
1255 2004-05-29 Seth Nickell <seth@gnome.org>
1258 * python/examples/example-client.py:
1259 * python/examples/example-service.py:
1260 * python/examples/list-system-services.py:
1262 Add SessionBus, SystemBus and ActivationBus classes
1263 so you don't need to know the special little BUS_TYPE
1266 2004-05-29 Havoc Pennington <hp@redhat.com>
1268 * bus/config-parser.c (process_test_valid_subdir): temporarily
1269 stop testing config parser OOM handling, since expat has issues
1270 http://freedesktop.org/pipermail/dbus/2004-May/001153.html
1272 * bus/dbus-daemon-1.1.in: change requested_reply to
1273 send_requested_reply/receive_requested_reply so we can send the
1274 replies, not just receive them.
1276 * bus/config-parser.c: parse the new
1277 send_requested_reply/receive_requested_reply
1279 * bus/policy.c (bus_client_policy_check_can_send): add
1280 requested_reply argument and use it
1282 * bus/bus.c (bus_context_check_security_policy): pass through
1283 requested_reply status to message send check
1285 * bus/system.conf.in: adapt to requested_reply change
1287 2004-05-28 Havoc Pennington <hp@redhat.com>
1289 * test/glib/test-service-glib.c (main): remove unused variable
1291 * glib/dbus-gidl.c (base_info_ref): fix a silly compiler warning
1293 * dbus/dbus-auth.h (enum): remove AUTHENTICATED_WITH_UNUSED_BYTES
1294 from the enum, no longer in use.
1296 * dbus/dbus-sysdeps.h: include config.h so DBUS_VA_COPY actually
1299 * dbus/dbus-message.c: add various _dbus_return_val_if_fail for
1300 whether error_name passed in is a valid error name.
1302 2004-05-28 John (J5) Palmieri <johnp@redhat.com>
1304 * dbus/dbus-message.c (dbus_message_get_args): Added support for
1305 OBJECT_PATH and OBJECT_PATH_ARRAY
1307 2004-05-28 Seth Nickell <seth@gnome.org>
1309 * python/examples/Makefile.am:
1311 Forget to add Makefile.am. Do not pass go.
1313 2004-05-28 Michael Meeks <michael@ximian.com>
1315 * glib/dbus-gvalue.c (dbus_gvalue_marshal, dbus_gvalue_demarshal):
1318 * dbus/dbus-string.c (_dbus_string_parse_basic_type): impl.
1320 * dbus/dbus-message.c (_dbus_message_iter_get_basic_type),
1321 (_dbus_message_iter_get_basic_type_array): impl.
1322 drastically simplify ~all relevant _get methods to use these.
1323 (_dbus_message_iter_append_basic_array),
1324 (dbus_message_iter_append_basic): impl
1325 drastically simplify ~all relevant _append methods to use these.
1327 * dbus/dbus-message-builder.c (parse_basic_type)
1328 (parse_basic_array, lookup_basic_type): impl.
1329 (_dbus_message_data_load): prune scads of duplicate /
1332 * dbus/dbus-marshal.c (_dbus_demarshal_basic_type_array)
1333 (_dbus_demarshal_basic_type): implement,
1334 (demarshal_and_validate_len/arg): beef up debug.
1335 (_dbus_marshal_basic_type, _dbus_marshal_basic_type_array): impl.
1337 2004-05-27 Seth Nickell <seth@gnome.org>
1340 * python/Makefile.am:
1342 Include the example python apps in the tarball.
1344 * python/examples/list-system-services.py
1346 Add a python new example that fetches the list of services
1347 from the system bus.
1349 2004-05-27 Seth Nickell <seth@gnome.org>
1352 * python/dbus_bindings.pyx.in:
1354 Fix failure to notify that a signal was not handled,
1355 resulted in hung functions.
1357 2004-05-25 Colin Walters <walters@redhat.com>
1359 * tools/dbus-monitor.c (main): Monitor all types of messages.
1361 2004-05-23 Owen Fraser-Green <owen@discobabe.net>
1363 * mono/Handler.cs, mono/Service.cs: Added UnregisterObject method
1364 which unregisters the object path and disposes the handler.
1366 2004-05-23 Kristian Høgsberg <krh@redhat.com>
1368 Patch from Timo Teräs <ext-timo.teras@nokia.com> (#614):
1370 * dbus/dbus-message.c (dbus_message_iter_get_args_valist): Swap
1371 operands to && so we call dbus_message_iter_next () for the last
1374 2004-05-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
1376 * dbus/dbus-object-tree.c
1377 (_dbus_object_tree_list_registered_unlock, lookup_subtree): return
1378 children even if the requested path isn't registered.
1379 (object_tree_test_iteration): test object_tree_list_registered.
1381 * configure.in: undefine HAVE_ABSTRACT_SOCKETS instead of defining
1384 2004-05-20 Kristian Høgsberg <krh@redhat.com>
1386 * doc/TODO: Remove resolved items.
1388 * bus/expirelist.h (struct BusExpireList): remove unused n_items
1391 * bus/connection.c (bus_connections_expect_reply): Enforce the
1392 per-connection limit on pending replies.
1394 Patch from Jon Trowbridge <trow@ximian.com>:
1396 * bus/main.c (setup_reload_pipe): Added. Creates a pipe and sets
1397 up a watch that triggers a config reload when one end of the pipe
1399 (signal_handler): Instead of doing the config reload in our SIGHUP
1400 handler, just write to the reload pipe and let the associated
1401 watch handle the reload when control returns to the main loop.
1403 * bus/driver.c (bus_driver_handle_reload_config): Added.
1404 Implements a ReloadConfig method for requesting a configuration
1405 file reload via the bus driver.
1407 2004-05-19 Owen Fraser-Green <owen@discobabe.net>
1409 * HACKING: Updated release instructions concerning the wiki page.
1411 2004-05-18 Kristian Høgsberg <krh@redhat.com>
1413 * dbus/dbus-auth.c (client_try_next_mechanism): Remove logic to
1414 filter against auth->allowed_mechs; we only add allowed mechs in
1415 record_mechanisms().
1417 * dbus/dbus-auth-script.c (_dbus_auth_script_run): Add an
1418 ALLOWED_MECHS to auth-script format so we can set the list of
1421 * data/auth/client-out-of-mechanisms.auth-script: New test to
1422 check client disconnects when it is out of mechanisms to try.
1424 * dbus/dbus-auth.c (process_command): Remove check for lines
1425 longer that 1 MB; we only buffer up maximum 16 kB.
1427 * dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
1428 dbus/dbus-auth-script.c, dbus/dbus-auth.c, dbus/dbus-auth.h:
1429 Remove auth state AUTHENTICATED_WITH_UNUSED_BYTES, instead always
1430 assume there might be unused bytes.
1432 * dbus/dbus-auth.c (_dbus_auth_do_work): Remove check for
1433 client-out-of-mechs, it is handled in process_reject(). Move check
1434 for max failures to send_rejected(), as it's a server-only thing.
1436 * dbus/dbus-auth.c: Factor out protocol reply code into functions
1437 send_auth(), send_data(), send_rejected(), send_error(),
1438 send_ok(), send_begin() and send_cancel().
1440 2004-05-17 Kristian Høgsberg <krh@redhat.com>
1442 Remove base64 encoding, replace with hex encoding. Original patch
1443 from trow@ximian.com, added error handling.
1445 * dbus/dbus-string.c (_dbus_string_base64_encode)
1446 (_dbus_string_base64_decode): Remove.
1447 (_dbus_string_hex_decode): Add end_return argument so we can
1448 distinguish between OOM and invalid hex encoding.
1449 (_dbus_string_test): Remove base64 tests and add test case for
1452 * dbus/dbus-keyring.c, dbus/dbus-auth-script.c, dbus/dbus-auth.c:
1453 Replace base64 with hex.
1455 * test/data/auth/invalid-hex-encoding.auth-script: New test case
1456 for invalid hex encoded data in auth protocol.
1458 2004-05-17 Olivier Andrieu <oliv__a@users.sourceforge.net>
1460 * dbus/dbus-connection.c (check_for_reply_unlocked): plug a memory
1463 2004-05-15 Owen Fraser-Green <owen@discobabe.net>
1465 * mono/dbus-sharp.dll.config.in: Added for GAC
1466 * mono/dbus-sharp.snk: Added for GAC
1467 * mono/Assembly.cs.in: Added for GAC
1468 * mono/Makefile.am: Changes for GAC installation
1469 * configure.in: Added refs for dbus-sharp.dll.config.in and
1470 Assembly.cs.in. More fixes for mono testing
1471 * mono/example/Makefile.am: Changed var to CSC
1472 * Makefile.am: Changed flag name to DBUS_USE_CSC
1474 2004-05-15 Owen Fraser-Green <owen@discobabe.net>
1476 * mono/Makefile.am: Added SUBDIRS for docs. Changed SUBDIRS order
1477 * mono/doc/*: Added documentation framework
1478 * configure.in: Added monodoc check
1479 * README: Added description of mono configure flags
1481 2004-05-11 John (J5) Palmieri <johnp@redhat.com>:
1483 * doc/dbus-specification.xml: Added a "Required" column to the
1484 header fields table and changed the "zero or more" verbage in
1485 the above paragraph to read "The header must contain the required
1486 named header fields and zero or more of the optional named header
1488 * test/data/invalid-messages/*.message: Added the required PATH
1489 named header field to the tests so that they don't fail on
1490 'Missing path field'
1492 2004-05-07 John (J5) Palmieri <johnp@redhat.com>
1494 * python/dbus-bindings.pyx.in: Stopped the bindings from trashing
1495 the stack by implicitly defining variable and parameter types and
1496 removing the hack of defining C pointers as python objects and later
1499 2004-05-02 Owen Fraser-Green <owen@discobabe.net>
1501 * mono/Makefile.am: Removed test-dbus-sharp.exe from all target
1503 2004-05-01 Owen Fraser-Green <owen@discobabe.net>
1505 * mono/DBusType/Dict.cs: Handle empty dicts
1506 * mono/DBusType/Array.cs: Handle empty arrays
1507 * mono/Arguments.cs: Handle empty arguments
1509 2004-04-30 Owen Fraser-Green <owen@discobabe.net>
1511 * dbus-sharp.pc.in: Modified to include include Libs and Requires
1514 2004-04-25 Kristian Høgsberg <krh@redhat.com>
1516 * test/data/valid-messages/standard-*.message: Update message
1517 test scripts to new header field names.
1519 2004-04-22 John (J5) Palmieri <johnp@redhat.com>
1521 * test/break-loader.c (randomly_do_n_things): tracked down buffer
1522 overflow to times_we_did_each_thing array which would chop off the
1523 first character of the failure_dir string. Increased the size of
1524 the array to 7 to reflect the number of random mutation functions
1527 2004-04-21 Kristian Høgsberg <krh@redhat.com>
1529 * dbus/dbus-server-unix.c (unix_finalize): Don't unref
1530 unix_server->watch here, it is unreffed in disconnect.
1531 (_dbus_server_new_for_tcp_socket): convert NULL host to
1532 "localhost" here so we don't append NULL to address.
1534 * dbus/dbus-server.c (_dbus_server_test): Add test case for
1535 various addresses, including tcp with no explicit host.
1537 2004-04-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
1539 * dbus/dbus-message.c (decode_header_data, decode_string_field):
1540 fix incorrect setting of .name_offset in the HeaderField (it was
1541 off by two bytes, positioned right after the name and typecode)
1543 * bus/bus.c (bus_context_new, bus_context_unref): test before
1544 calling dbus_server_free_data_slot and _dbus_user_database_unref
1545 in case of an error.
1547 * tools/Makefile.am: add $(DBUS_GLIB_TOOL_LIBS), xml libs needed
1550 2004-04-19 Kristian Høgsberg <krh@redhat.com>
1552 * dbus/dbus-transport-unix.c (unix_do_iteration): Rewrite to use
1553 _dbus_poll() instead of select().
1555 2004-04-15 Jon Trowbridge <trow@ximian.com>
1557 * bus/main.c (signal_handler): Reload the configuration files
1559 (main): Set up our SIGHUP handler.
1561 * bus/bus.c (struct BusContext): Store the config file, user and
1562 fork flag in the BusContext.
1563 (process_config_first_time_only): Added. Contains the code
1564 (previously in bus_context_new) for setting up the BusContext from
1565 the BusConfigParser that should only be run the first time the
1566 config files are read.
1567 (process_config_every_time): Added. Contains the code (previously
1568 in bus_context_new) for setting up the BusContext from the
1569 BusConfigParser that should be run every time the config files are
1571 (load_config): Added. Builds a BusConfigParser from the config
1572 files and passes the resulting structure off to
1573 process_config_first_time_only (assuming this is the first time)
1574 and process_config_every_time.
1575 (bus_context_new): All of the config-related code has been moved
1576 to process_config_first_time_only and process_config_every_time.
1577 Now this function just does the non-config-related initializations
1578 and calls load_config.
1579 (bus_context_reload_config): Added.
1581 2004-04-15 Olivier Andrieu <oliv__a@users.sourceforge.net>
1583 * bus/driver.c (bus_driver_handle_get_service_owner):
1584 implement a GetServiceOwner method.
1585 * doc/dbus-specification.xml: document it.
1586 * dbus/dbus-errors.h: add a 'ServiceHasNoOwner' error.
1588 * glib/dbus-gproxy.c (dbus_gproxy_new_for_service_owner):
1589 implement, using the bus GetServiceOwner method.
1591 * test/glib/test-dbus-glib.c:
1592 use dbus_gproxy_new_for_service_owner so that we can receive the
1595 2004-04-15 John (J5) Palmieri <johnp@redhat.com>
1597 * dbus/dbus-internals.c, dbus/dbus-message-builder.c,
1598 dbus/dbus-message.c, dbus/dbus-protocol.h
1599 (DBUS_HEADER_FIELD_SERVICE): renamed DBUS_HEADER_FIELD_DESTINATION
1601 * dbus/dbus-internals.c, dbus/dbus-message-builder.c,
1602 dbus/dbus-message.c, dbus/dbus-protocol.h
1603 (DBUS_HEADER_FIELD_SENDER_SERVICE): renamed DBUS_HEADER_FIELD_SENDER
1605 * dbus/dbus-internals.c (_dbus_header_field_to_string):
1606 DBUS_HEADER_FIELD_DESTINATION resolves to "destination"
1607 DBUS_HEADER_FIELD_SENDER resolves to "sender"
1609 * doc/dbus-specification.xml (Header Fields Table):
1610 s/SERVICE/DESTINATION
1611 s/SENDER_SERVICE/SENDER
1614 2004-04-14 Olivier Andrieu <oliv__a@users.sourceforge.net>
1616 * test/glib/test-dbus-glib.c (timed_exit): fail the test after
1619 2004-04-13 Michael Meeks <michael@ximian.com>
1621 * glib/dbus-gobject.c (handle_introspect): split out
1622 (introspect_properties): this.
1623 (handle_introspect): implement this.
1625 * test/glib/Makefile.am: use the absolute path so the bus
1626 daemon's chdir ("/") doesn't kill us dead.
1628 * configure.in: subst ABSOLUTE_TOP_BUILDDIR
1630 2004-04-12 Jon Trowbridge <trow@ximian.com>
1632 * bus/config-parser.c (struct BusConfigParser): Added
1633 included_files field.
1634 (seen_include): Added. Checks whether or not a file has already
1635 been included by any parent BusConfigParser.
1636 (bus_config_parser_new): Copy the parent's included_files.
1637 (include_file): Track which files have been included, and fail on
1638 circular inclusions.
1639 (process_test_valid_subdir): Changed printf to report if we are
1640 testing valid or invalid conf files.
1641 (all_are_equiv): Changed printf to be a bit clearer about
1642 what we are actually doing.
1643 (bus_config_parser_test): Test invalid configuration files.
1645 2004-04-09 Jon Trowbridge <trow@ximian.com>
1647 * bus/config-parser.c (bus_config_parser_new): Added a 'parent'
1648 argument. If non-null, the newly-constructed BusConfigParser will
1649 be initialized with the parent's BusLimits instead of the default
1651 (include_file): When including a config file, pass in
1652 the current parser as the parent and then copy the BusLimits
1653 from the included BusConfigParser pack to the current parser.
1654 (process_test_valid_subdir): Renamed from process_test_subdir.
1655 (process_test_equiv_subdir): Added. Walks through a directory,
1656 descending into each subdirectory and loading the config files
1657 it finds there. If any subdirectory contains two config files
1658 that don't produce identical BusConfigParser structs, fail.
1659 For now, the BusConfigParser's BusPolicies are not compared.
1660 (bus_config_parser_test): Call both process_test_valid_subdir and
1661 process_test_equiv_subdir.
1663 * bus/config-loader-libxml.c (bus_config_load): Take a parent
1664 argument and pass it along to the call to bus_config_parser_new.
1665 Also made a few small changes to allow this code to compile.
1667 * bus/config-loader-expat.c (bus_config_load): Take a parent
1668 argument and pass it along to the call to bus_config_parser_new.
1670 * bus/bus.c (bus_context_new): Load the config file
1671 with a NULL parent argument.
1673 2004-03-29 Michael Meeks <michael@ximian.com>
1675 * glib/dbus-gobject.c (introspect_properties): split
1676 out, fix mangled 'while' flow control.
1677 (introspect_signals): implement.
1678 (handle_introspect): update.
1680 2004-03-29 Michael Meeks <michael@ximian.com>
1682 * glib/dbus-gobject.c (set_object_property): split out /
1683 re-work, use the property type, and not the message type(!)
1684 (get_object_property): ditto.
1686 * glib/dbus-gvalue.c (dbus_gvalue_demarshal),
1687 (dbus_gvalue_marshal): make this code re-usable, needed
1688 for signals too, also on both proxy and server side.
1689 Re-write for more efficiency / readability.
1691 2004-03-29 Michael Meeks <michael@ximian.com>
1693 * dbus/dbus-message.c
1694 (dbus_message_new_error_printf): impl.
1696 * dbus/dbus-connection.c
1697 (dbus_connection_unregister_object_path): fix warning.
1699 * configure.in: fix no-mono-installed situation.
1701 2004-03-27 Havoc Pennington <hp@redhat.com>
1703 Patch from Timo Teräs:
1705 * tools/dbus-send.c (main): if --print-reply, assume type is
1706 method call; support boolean type args
1708 * dbus/dbus-connection.c (dbus_connection_send_with_reply): fix a
1709 bunch of memleak and logic bugs
1711 2004-03-23 Owen Fraser-Green <owen@discobabe.net>
1713 * mono/Arguments.cs:
1714 * mono/Introspector.cs:
1716 * mono/InterfaceProxy.cs:
1718 * mono/ProxyBuilder.cs:
1720 Added InterfaceProxy class to avoid building proxies for every
1724 * dbus-message.c (dbus_message_append_args_valist)
1725 (dbus_message_iter_get_object_path)
1726 (dbus_message_iter_get_object_path_array)
1727 (dbus_message_iter_append_object_path)
1728 (dbus_message_iter_append_object_path_array):
1729 Added object_path iter functions to handle OBJECT_PATH arguments
1731 2004-03-23 Owen Fraser-Green <owen@discobabe.net>
1733 First checkin of mono bindings.
1736 Build stuff for the bindings
1737 * dbus-sharp.pc.in: Added for pkgconfig
1739 2004-03-21 Havoc Pennington <hp@redhat.com>
1741 * test/test-service.c (main): remove debug spew
1743 2004-03-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
1745 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): accept empty
1748 * dbus/dbus-message.h, bus/dbus-message.c (dbus_message_iter_init)
1749 (dbus_message_iter_init_array_iterator)
1750 (dbus_message_iter_init_dict_iterator): return a dbus_bool_t to
1751 indicate whether the iterator is empty
1753 * dbus/dbus-pending-call.c, dbus/dbus-server.c: silence compiler
1756 2004-03-19 Havoc Pennington <hp@redhat.com>
1758 * NEWS: 0.21 updates
1760 * configure.in: 0.21
1762 * doc/Makefile.am: add all XMLTO usage to DBUS_XML_DOCS_ENABLED
1764 * python/Makefile.am: change to avoid dist of dbus_bindings.c so
1765 you don't need pyrex to make dist
1767 * qt/Makefile.am (libdbus_qt_1_la_SOURCES): add integrator.h to
1770 2004-03-18 Richard Hult <richard@imendio.com>
1772 * dbus/dbus-message.c (dbus_message_get_auto_activation)
1773 (dbus_message_set_auto_activation): Add doxygen docs.
1775 2004-03-16 Richard Hult <richard@imendio.com>
1777 * bus/activation.c: (bus_activation_service_created),
1778 (bus_activation_send_pending_auto_activation_messages),
1779 (bus_activation_activate_service):
1781 * bus/dispatch.c: (bus_dispatch),
1782 (check_nonexistent_service_auto_activation),
1783 (check_service_auto_activated),
1784 (check_segfault_service_auto_activation),
1785 (check_existent_service_auto_activation), (bus_dispatch_test):
1786 * bus/driver.c: (bus_driver_handle_activate_service):
1787 * bus/services.c: (bus_registry_acquire_service):
1788 * dbus/dbus-message.c: (dbus_message_set_auto_activation),
1789 (dbus_message_get_auto_activation):
1790 * dbus/dbus-message.h:
1791 * dbus/dbus-protocol.h: Implement auto-activation.
1793 * doc/dbus-specification.xml: Add auto-activation to the spec.
1795 2004-03-12 Olivier Andrieu <oliv__a@users.sourceforge.net>
1797 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos):
1798 fix a bug with CUSTOM types.
1800 * dbus/dbus-message.c (message_iter_test, _dbus_message_test): add
1801 a unit test for this bug (used to fail).
1803 2004-03-12 Mikael Hallendal <micke@imendio.com>
1806 (babysitter_watch_callback): notify all pending activations waiting for
1807 the same exec that the activation failed.
1808 (bus_activation_activate_service): shortcut the activation if we
1809 already waiting for the same executable to start up.
1811 2004-03-12 Mikael Hallendal <micke@imendio.com>
1814 - Added service file reloading.
1815 Each service files directory is kept in an hash table in
1816 BusActivation and each BusActivationEntry knows what .service-file it
1817 was read from. So when you try to activate a service the bus will
1818 check if it's been updated, removed or if new .service-files has
1820 - Test code at the bottom for the service file reloading.
1821 * bus/test-main.c: (main):
1823 - added service reloading test.
1824 * dbus/dbus-sysdeps.c:
1825 * dbus/dbus-sysdeps.h: (_dbus_delete_directory): Added.
1827 2004-03-08 Michael Meeks <michael@ximian.com>
1829 * dbus/dbus-connection.c (_dbus_connection_block_for_reply):
1830 bail immediately if disconnected, to avoid busy loop.
1832 * dbus/dbus-message.c (dbus_message_iter_get_args_valist):
1833 cleanup cut/paste/inefficiency.
1835 2004-03-01 David Zeuthen <david@fubar.dk>
1837 * dbus/dbus-string.c (_dbus_string_append_printf_valist): Fix a
1838 bug where args were used twice. This bug resulted in a segfault
1839 on a Debian/PPC system when starting the messagebus daemon. Include
1840 dbus-sysdeps.h for DBUS_VA_COPY
1842 * dbus/dbus-sysdeps.h: Define DBUS_VA_COPY if neccessary. From GLib
1844 * configure.in: Check for va_copy; define DBUS_VA_COPY to the
1845 appropriate va_copy implementation. From GLib
1847 2004-02-24 Joe Shaw <joe@ximian.com>
1849 * bus/services.c (bus_registry_acquire_service): We need to pass
1850 in the service name to dbus_set_error() to prevent a crash.
1852 2003-12-26 Anders Carlsson <andersca@gnome.org>
1854 * AUTHORS: Reveal my True identity.
1856 2003-12-17 Mikael Hallendal <micke@imendio.com>
1858 * dbus/dbus-message.c: (dbus_message_append_args_valist):
1859 - Added case for DBUS_TYPE_BYTE, patch from Johan Hedberg.
1861 2003-12-13 Mikael Hallendal <micke@imendio.com>
1863 * doc/TODO: Added not about better error check of configuration files.
1865 2003-12-02 Richard Hult <richard@imendio.com>
1867 * Update AFL version to 2.0 throughout the source files to reflect
1868 the update that was done a while ago.
1870 2003-12-02 Richard Hult <richard@imendio.com>
1872 * dbus/dbus-message.c (dbus_message_iter_append_dict): Set
1873 wrote_dict_key to FALSE on the iter that the dict is appended to,
1874 just like when appending other types. Fixes a bug where a dict
1875 couldn't be put inside a dict.
1876 (dbus_message_iter_append_dict_key): Fix typo in warning message.
1877 (message_iter_test, _dbus_message_test): Add test case for dict
1880 2003-12-01 David Zeuthen <david@fubar.dk>
1882 * python/dbus.py: Add the actual message when calling the reciever
1883 of a signal such that parameters can be inspected. Add the method
1884 remove_signal_receiver
1886 2003-11-26 Mikael Hallendal <micke@imendio.com>
1890 * glib/*.[ch]: Made ref functions return the pointer
1892 2003-11-25 Zack Rusin <zack@kde.org>
1894 * qt/integrator.h, qt/integrator.cpp: Adding handling of DBusServer,
1896 * qt/server.h, qt/server.cpp, qt/Makefile.am: Adding DBusServer
1899 * qt/connection.h, qt/connection.cpp: Adjusting to changes in
1900 the Integrator and to better fit with the server,
1902 2003-11-24 Zack Rusin <zack@kde.org>
1904 * qt/connection.h, qt/connection.cpp: removing initDbus method since
1905 the integrator handles it now
1907 * qt/integrator.h, qt/integrator.cpp: reworking handling of timeouts,
1908 since QTimer wasn't really meant to be used the way DBusTimeout is
1910 2003-11-24 Zack Rusin <zack@kde.org>
1912 * qt/integrator.h, qt/integrator.cpp, Makefile.am: Adding
1913 Integrator class which integrates D-BUS with the Qt event loop,
1915 * qt/connection.h, qt/connection.cpp: Move all the code which
1916 was dealing with D-BUS integration to the Integrator class,
1917 and start using Integrator,
1919 2003-11-23 Zack Rusin <zack@kde.org>
1921 * qt/connection.h, qt/connection.cpp: Adding the DBusConnection
1924 * qt/message.h, qt/message.cpp: updating to the current D-BUS api,
1925 switching namespaces to DBusQt, reworking the class,
1927 * Makefile.cvs: switching dependencies so that it matches KDE
1930 * qt/Makefile.am: adding connection.{h,cpp} and message.{h,cpp} to
1933 2003-11-19 Havoc Pennington <hp@redhat.com>
1937 * configure.in: bump version to 0.20
1939 * configure.in (have_qt): add yet another place to look for qt
1940 (someone hand trolltech a .pc file...)
1942 2003-11-01 Havoc Pennington <hp@redhat.com>
1944 * doc/dbus-specification.xml: add state machine docs on the auth
1945 protocol; just a first draft, I'm sure it's wrong.
1947 2003-10-28 David Zeuthen <david@fubar.dk>
1949 * python/dbus_bindings.pyx.in: add get_dict to handle dictionaries
1950 return types. Fixup TYPE_* to reflect changes in dbus/dbus-protocol.h
1952 2003-10-28 Havoc Pennington <hp@redhat.com>
1954 * dbus/dbus-message.c (get_next_field): delete unused function
1956 2003-10-28 Havoc Pennington <hp@redhat.com>
1958 * bus/expirelist.c (do_expiration_with_current_time): detect
1959 failure of the expire_func due to OOM
1961 * bus/connection.c (bus_pending_reply_expired): return FALSE on OOM
1963 * bus/dispatch.c (check_send_exit_to_service): fix to handle the
1964 NoReply error that's now created by the bus when the service exits
1966 2003-10-28 Havoc Pennington <hp@redhat.com>
1968 * dbus/dbus-message.c (_dbus_message_test): enable and fix the
1969 tests for set_path, set_interface, set_member, etc.
1971 * dbus/dbus-string.c (_dbus_string_insert_bytes): allow 0 bytes
1973 * dbus/dbus-message.c (set_string_field): always just delete and
1974 re-append the field; accept NULL for deletion
1975 (re_align_fields_recurse): reimplement
1977 2003-10-26 Havoc Pennington <hp@redhat.com>
1979 * dbus/dbus-connection.c: fix docs to properly describe the
1980 disconnected message
1981 (_dbus_connection_notify_disconnected): remove this function;
1982 we can't synchronously add the disconnected message, we have to
1983 do it after we've queued any remaining real messages
1984 (_dbus_connection_get_dispatch_status_unlocked): queue the
1985 disconnect message only if the transport has finished queueing all
1986 its real messages and is disconnected.
1987 (dbus_connection_disconnect): update the dispatch status here
1989 2003-10-22 Havoc Pennington <hp@redhat.com>
1991 * bus/bus.c (bus_context_check_security_policy): fix up assertion
1993 * bus/connection.c (bus_transaction_send_from_driver): set the
1994 destination to the connection's base service
1996 2003-10-20 Havoc Pennington <hp@redhat.com>
1998 hmm, make check is currently not passing.
2000 * doc/dbus-specification.xml: add requirement that custom type
2001 names follow the same rules as interface names.
2003 * dbus/dbus-protocol.h: change some of the byte codes, to avoid
2004 duplication and allow 'c' to be 'custom'; dict is now 'm' for
2007 * doc/dbus-specification.xml: update type codes to match
2008 dbus-protocol.h, using the ASCII byte values. Rename type NAMED to
2009 CUSTOM. Add type OBJECT_PATH to the spec.
2011 2003-10-17 Havoc Pennington <hp@redhat.com>
2013 * bus/driver.c (create_unique_client_name): use "." as separator
2014 in base service names instead of '-'
2016 * dbus/dbus-string.c (_dbus_string_get_byte): allow getting nul
2017 byte at the end of the string
2019 * dbus/dbus-internals.h (_DBUS_LIKELY, _DBUS_UNLIKELY): add
2020 optimization macros since string validation seems to be a slow
2023 * doc/dbus-specification.xml: restrict valid
2024 service/interface/member/error names. Add test suite code for the
2027 * dbus/dbus-string.c: limit service/interface/member/error names
2030 * dbus/dbus-connection.c (dbus_connection_dispatch): add missing
2031 format arg to verbose spew
2033 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): if not out of
2034 memory, return instead of g_error
2036 * test/test-service.c (path_message_func): support emitting a
2039 * dbus/dbus-bus.c (init_connections_unlocked): only fill in
2040 activation bus type if DBUS_BUS_ACTIVATION was set; default to
2041 assuming the activation bus was the session bus so that services
2042 started manually will still register.
2043 (init_connections_unlocked): fix so that in OOM situation we get
2044 the same semantics when retrying the function
2046 * test/test-service.c (main): change to use path registration, to
2047 test those codepaths; register with DBUS_BUS_ACTIVATION rather
2048 than DBUS_BUS_SESSION
2050 2003-10-16 Havoc Pennington <hp@redhat.com>
2052 * glib/dbus-gtest-main.c: bracket with #ifdef DBUS_BUILD_TESTS
2054 * Makefile.am (GCOV_DIRS): remove "test", we don't care about test
2055 coverage of the tests
2056 (coverage-report.txt): don't move the .da and .bbg files around
2058 2003-10-16 Havoc Pennington <hp@redhat.com>
2060 * bus/bus.c (struct BusContext): remove struct field I didn't mean
2063 2003-10-16 Havoc Pennington <hp@redhat.com>
2065 * bus/connection.c (bus_pending_reply_expired): either cancel or
2067 (bus_connections_check_reply): use unlink, not remove_link, as we
2068 don't want to free the link; fixes double free mess
2070 * dbus/dbus-pending-call.c (dbus_pending_call_block): fix in case
2071 where no reply was received
2073 * dbus/dbus-connection.c (_dbus_pending_call_complete_and_unlock):
2076 * bus/signals.c (match_rule_matches): add special cases for the
2077 bus driver, so you can match on sender/destination for it.
2079 * dbus/dbus-sysdeps.c (_dbus_abort): print backtrace if
2080 DBUS_PRINT_BACKTRACE is set
2082 * dbus/dbus-internals.c: add pid to assertion failure messages
2084 * dbus/dbus-connection.c: add message type code to the debug spew
2086 * glib/dbus-gproxy.c (gproxy_get_match_rule): match rules want
2087 sender=foo not service=foo
2089 * dbus/dbus-bus.c (dbus_bus_get): if the activation bus is the
2090 session bus but DBUS_SESSION_BUS_ADDRESS isn't set, use
2091 DBUS_ACTIVATION_ADDRESS instead
2093 * bus/activation.c: set DBUS_SESSION_BUS_ADDRESS,
2094 DBUS_SYSTEM_BUS_ADDRESS if appropriate
2096 * bus/bus.c (bus_context_new): handle OOM copying bus type into
2099 * dbus/dbus-message.c (dbus_message_iter_get_object_path): new function
2100 (dbus_message_iter_get_object_path_array): new function (half
2101 finished, disabled for the moment)
2103 * glib/dbus-gproxy.c (dbus_gproxy_end_call): properly handle
2104 DBUS_MESSAGE_TYPE_ERROR
2106 * tools/dbus-launch.c (babysit): support DBUS_DEBUG_OUTPUT to
2107 avoid redirecting stderr to /dev/null
2108 (babysit): close stdin if not doing the "exit_with_session" thing
2110 * dbus/dbus-sysdeps.c (_dbus_become_daemon): delete some leftover
2111 debug code; change DBUS_DEBUG_OUTPUT to only enable stderr, not
2112 stdout/stdin, so things don't get confused
2114 * bus/system.conf.in: fix to allow replies, I modified .conf
2115 instead of .conf.in again.
2117 2003-10-14 David Zeuthen <david@fubar.dk>
2119 * python/dbus_bindings.pyx.in (MessageIter.get): fixed typo in
2120 argtype to arg_type when raising unknown arg type exception.
2121 Changed type list to reflect the changes in dbus-protocol.h so
2122 the bindings actually work.
2124 2003-10-14 Havoc Pennington <hp@redhat.com>
2126 * test/decode-gcov.c: support gcc 3.3 also, though gcc 3.3 seems
2127 to have a bug keeping it from outputting the .da files sometimes
2128 (string_get_string): don't append garbage nul bytes to the string.
2130 2003-10-15 Seth Nickell <seth@gnome.org>
2132 * python/Makefile.am:
2134 Include dbus_h_wrapper.h in the dist tarball.
2136 2003-10-14 Havoc Pennington <hp@redhat.com>
2138 * bus/bus.c (bus_context_check_security_policy): revamp this to
2139 work more sanely with new policy-based requested reply setup
2141 * bus/connection.c (bus_transaction_send_from_driver): set bus
2142 driver messages as no reply
2144 * bus/policy.c (bus_client_policy_check_can_receive): handle a
2145 requested_reply attribute on allow/deny rules
2147 * bus/system.conf: add <allow requested_reply="true"/>
2149 * bus/driver.c (bus_driver_handle_message): fix check for replies
2150 sent to the bus driver, which was backward. How did this ever work
2151 at all though? I think I'm missing something.
2153 * dbus/dbus-message.c (decode_header_data): require error and
2154 method return messages to have a reply serial field to be valid
2155 (_dbus_message_loader_queue_messages): break up this function;
2156 validate that reply serial and plain serial are nonzero;
2157 clean up the OOM/error handling.
2158 (get_uint_field): don't return -1 from this
2159 (dbus_message_create_header): fix signed/unsigned bug
2161 * bus/connection.c (bus_connections_expect_reply): save serial of
2162 the incoming message, not reply serial
2164 2003-10-14 Havoc Pennington <hp@redhat.com>
2166 * bus/connection.c: implement pending reply tracking using
2169 * bus/bus.c (bus_context_check_security_policy): verify that a
2170 reply is pending in order to allow a reply to be sent. Deny
2171 messages of unknown type.
2173 * bus/dbus-daemon-1.1.in: update to mention new resource limits
2175 * bus/bus.c (bus_context_get_max_replies_per_connection): new
2176 (bus_context_get_reply_timeout): new
2178 2003-10-13 Seth Nickell <seth@gnome.org>
2180 * python/Makefile.am:
2182 Pass "make distcheck": remove a couple files from DIST_FILES
2183 that weren't included in the final version.
2185 2003-10-12 Havoc Pennington <hp@pobox.com>
2187 Added test code that 1) starts an actual bus daemon and 2) uses
2188 DBusGProxy; fixed bugs that were revealed by the test. Lots
2189 more testing possible, but this is the basic framework.
2191 * glib/dbus-gproxy.c (dbus_gproxy_manager_unregister): remove
2192 empty proxy lists from the proxy list hash
2194 * dbus/dbus-message.c (dbus_message_iter_get_args_valist): add a
2195 couple of return_if_fail checks
2197 * dbus/dbus-pending-call.c (_dbus_pending_call_new): use dbus_new0
2198 to allocate, so everything is cleared to NULL as it should be.
2200 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): pass
2201 source as data to dbus_connection_set_timeout_functions() as the
2202 timeout functions expected
2204 * test/glib/run-test.sh: add a little script to start up a message
2205 bus and run tests using it
2207 * tools/dbus-launch.1: updates
2209 * tools/dbus-launch.c (main): add --config-file option
2211 * tools/dbus-launch.c (main): remove confusing else if (runprog)
2212 that could never be reached.
2214 * dbus/dbus-message.c (dbus_message_new_method_return)
2215 (dbus_message_new_error, dbus_message_new_signal): set the
2216 no-reply-expected flag on all these. Redundant, but may
2217 as well be consistent.
2219 2003-10-11 Havoc Pennington <hp@pobox.com>
2221 * test/decode-gcov.c (function_solve_graph): make broken block
2222 graph a nonfatal error since it seems to be broken. Need to debug
2225 * dbus/dbus-marshal.c (_dbus_type_is_valid): new function since we
2226 can't just check type > INVALID < LAST anymore
2228 * dbus/dbus-message.c (dbus_message_get_signature): new function
2229 (dbus_message_has_signature): new function
2230 (struct DBusMessage): add signature field (right now it isn't sent
2231 over the wire, just generated on the fly)
2232 (dbus_message_copy): copy the signature, and init strings to
2233 proper length to avoid some reallocs
2234 (dbus_message_iter_init_array_iterator): return void, since it
2236 (dbus_message_iter_init_dict_iterator): return void since it can't fail
2237 (_dbus_message_loader_queue_messages): add silly temporary hack to
2238 fill in message->signature on load
2240 * dbus/dbus-protocol.h: change DBUS_TYPE_* values to be ASCII
2241 characters, so they are relatively human-readable.
2243 2003-10-11 Havoc Pennington <hp@pobox.com>
2245 * dbus/dbus-message.c (_dbus_message_test): add more test
2246 coverage, but #if 0 for now since they uncover a bug
2247 not fixed yet; I think in re_align_field_recurse()
2248 (re_align_field_recurse): add FIXME about broken assertion
2250 * dbus/dbus-sysdeps.c (_dbus_sysdeps_test): add more test coverage
2252 * bus/connection.c: share a couple code bits with expirelist.c
2254 * bus/expirelist.h, bus/expirelist.c: implement a generic
2255 expire-items-after-N-seconds facility, was going to share between
2256 expiring connections and replies, decided not to use for expiring
2257 connections for now.
2259 * COPYING: include AFL 2.0 (still need to change all the file headers)
2261 2003-10-09 Havoc Pennington <hp@redhat.com>
2263 * configure.in: define DBUS_HAVE_GCC33_GCOV if we have
2264 gcc 3.3. Not that we do anything about it yet.
2266 * bus/signals.c (bus_match_rule_parse): impose max length on the
2269 * dbus/dbus-protocol.h: add DBUS_MAXIMUM_MATCH_RULE_LENGTH
2271 2003-10-09 Havoc Pennington <hp@redhat.com>
2273 Make matching rules theoretically work (add parser).
2275 * bus/bus.c (bus_context_check_security_policy): fix up to handle
2276 the case where destination is explicitly specified as bus driver
2277 and someone else is eavesdropping.
2279 * bus/policy.c (bus_client_policy_check_can_receive): fix up
2280 definition of eavesdropping and assertion
2282 * tools/dbus-send.c (main): use dbus_message_type_from_string
2284 * bus/signals.c (bus_match_rule_parse): implement
2286 * dbus/dbus-message.c (dbus_message_type_from_string): new
2288 * dbus/dbus-errors.h (DBUS_ERROR_MATCH_RULE_INVALID): add
2290 2003-10-02 Havoc Pennington <hp@pobox.com>
2292 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): rename from
2293 dbus_gproxy_oneway_call
2295 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main)
2296 (dbus_server_setup_with_g_main): fix to allow calling them more
2297 than once on the same args
2298 (dbus_bus_get_with_g_main): new function
2300 2003-10-02 Havoc Pennington <hp@redhat.com>
2302 * doc/dbus-tutorial.xml: write some stuff
2304 2003-09-29 Havoc Pennington <hp@pobox.com>
2306 * configure.in: split checks for Doxygen from XML docs, check for
2309 * doc/Makefile.am: XML-ify all the docs, and add a blank
2312 2003-09-29 Havoc Pennington <hp@pobox.com>
2314 * Merge dbus-object-names branch. To see the entire patch
2315 do cvs diff -r DBUS_OBJECT_NAMES_BRANCHPOINT -r dbus-object-names,
2316 it's huuuuge though.
2317 To revert, I tagged DBUS_BEFORE_OBJECT_NAMES_MERGE.
2319 2003-09-28 Havoc Pennington <hp@pobox.com>
2321 * HACKING: update to reflect new server
2323 2003-09-26 Seth Nickell <seth@gnome.org>
2326 * python/examples/example-signals.py:
2328 Start implementing some notions of signals. The API
2329 is really terrible, but they sort of work (with the
2330 exception of being able to filter by service, and to
2331 transmit signals *as* a particular service). Need to
2332 figure out how to make messages come from the service
2335 * python/dbus_bindings.pyx.in:
2337 Removed duplicate message_handler callbacks.
2339 2003-09-25 Havoc Pennington <hp@redhat.com>
2341 * bus/session.conf.in: fix my mess
2343 2003-09-25 Havoc Pennington <hp@pobox.com>
2345 * bus/session.conf.in: fix security policy, reported by Seth Nickell
2347 2003-09-25 Seth Nickell <seth@gnome.org>
2349 * python/examples/example-service.py:
2351 Johan notices complete wrong code in example-service, but
2352 completely wrong in a way that works exactly the same (!).
2353 Johan is confused, how could this possibly work? Example
2354 code fails to serve purpose of making things clear.
2357 2003-09-25 Mark McLoughlin <mark@skynet.ie>
2359 * doc/dbus-specification.sgml: don't require header fields
2360 to be 4-byte aligned and specify that fields should be
2361 distinguished from padding by the fact that zero is not
2364 * doc/TODO: remove re-alignment item and add item to doc
2365 the OBJECT_PATH type.
2367 * dbus/dbus-message.c:
2368 (HeaderField): rename the original member to value_offset
2369 and introduce a name_offset member to keep track of where
2370 the field actually begins.
2371 (adjust_field_offsets): remove.
2372 (append_int_field), (append_uint_field),
2373 (append_string_field): don't align the start of the header
2374 field to a 4-byte boundary.
2375 (get_next_field): impl finding the next marhsalled field
2376 after a given field.
2377 (re_align_field_recurse): impl re-aligning a number of
2378 already marshalled fields.
2379 (delete_field): impl deleting a field of any type and
2380 re-aligning any following fields.
2381 (delete_int_or_uint_field), (delete_string_field): remove.
2382 (set_int_field), (set_uint_field): no need to re-check
2383 that we have the correct type for the field.
2384 (set_string_field): ditto and impl re-aligning any
2386 (decode_header_data): update to take into account that
2387 the fields aren't 4-byte aligned any more and the new
2388 way to distinguish padding from header fields. Also,
2389 don't exit when there is too much header padding.
2390 (process_test_subdir): print the directory.
2391 (_dbus_message_test): add test to make sure a following
2392 field is re-aligned correctly after field deletion.
2394 * dbus/dbus-string.[ch]:
2395 (_dbus_string_insert_bytes): rename from insert_byte and
2396 allow the insert of multiple bytes.
2397 (_dbus_string_test): test inserting multiple bytes.
2399 * dbus/dbus-marshal.c: (_dbus_marshal_set_string): add
2400 warning note to docs about having to re-align any
2401 marshalled values following the string.
2403 * dbus/dbus-message-builder.c:
2404 (append_string_field), (_dbus_message_data_load):
2405 don't align the header field.
2407 * dbus/dbus-auth.c: (process_test_subdir): print the
2410 * test/break-loader.c: (randomly_add_one_byte): upd. for
2413 * test/data/invalid-messages/bad-header-field-alignment.message:
2416 * test/data/valid-messages/unknown-header-field.message: shove
2417 a dict in the unknown field.
2419 2003-09-25 Seth Nickell <seth@gnome.org>
2422 * python/dbus_bindings.pyx.in:
2424 Handle return values.
2426 * python/examples/example-client.py:
2427 * python/examples/example-service.py:
2429 Pass back return values from the service to the client.
2431 2003-09-24 Seth Nickell <seth@gnome.org>
2435 Connect Object methods (when you are sharing an object) up... pass
2436 in a list of methods to be shared. Sharing all the methods just
2437 worked out too weird. You can now create nice Services over the
2440 * python/dbus_bindings.pyx.in:
2442 Keep references to user_data tuples passed into C functions so
2443 Python doesn't garbage collect on us.
2445 Implement MethodReturn and Error subclasses of Message for creating
2446 DBusMessage's of those types.
2448 * python/examples/example-client.py:
2449 * python/examples/example-service.py:
2451 Simple example code showing both how create DBus services and objects,
2452 and how to use them.
2454 2003-09-23 Havoc Pennington <hp@pobox.com>
2456 * glib/dbus-gproxy.c (dbus_gproxy_manager_filter): implement
2458 2003-09-23 Havoc Pennington <hp@redhat.com>
2460 * glib/dbus-gproxy.c (dbus_gproxy_connect_signal): implement
2461 (dbus_gproxy_disconnect_signal): implement
2462 (dbus_gproxy_manager_remove_signal_match): implement
2463 (dbus_gproxy_manager_add_signal_match): implement
2464 (dbus_gproxy_oneway_call): implement
2466 2003-09-23 Havoc Pennington <hp@pobox.com>
2468 * glib/dbus-gproxy.c (struct DBusGProxy): convert to a GObject
2469 subclass. This means dropping the transparent thread safety of the
2470 proxy; you now need a separate proxy per-thread, or your own
2471 locking on the proxy. Probably right anyway.
2472 (dbus_gproxy_ref, dbus_gproxy_unref): nuke, just use g_object_ref
2474 2003-09-22 Havoc Pennington <hp@redhat.com>
2476 * glib/dbus-gproxy.c (dbus_gproxy_manager_get): implement
2478 2003-09-21 Seth Nickell <seth@gnome.org>
2480 First checkin of the Python bindings.
2482 * python/.cvsignore:
2483 * python/Makefile.am:
2484 * python/dbus_bindings.pyx.in:
2485 * python/dbus_h_wrapper.h:
2487 Pieces for Pyrex to operate on, building a dbus_bindings.so
2488 python module for low-level access to the DBus APIs.
2492 High-level Python module for accessing DBus objects.
2497 Build stuff for the python bindings.
2501 Extra macro needed for finding the Python C header files.
2503 2003-09-21 Havoc Pennington <hp@pobox.com>
2505 * glib/dbus-gproxy.c (dbus_gproxy_manager_new): start
2506 implementing the proxy manager, didn't get very far.
2508 * dbus/dbus-bus.c (dbus_bus_add_match): new
2509 (dbus_bus_remove_match): new
2511 * glib/dbus-gproxy.c (dbus_gproxy_new_for_service): add a
2512 path_name argument; adjust the other not-yet-implemented
2513 gproxy constructors to be what I think they should be.
2515 2003-09-21 Havoc Pennington <hp@pobox.com>
2517 * dbus/dbus-bus.c (dbus_bus_get): set exit_on_disconnect to TRUE
2518 by default for message bus connections.
2520 * dbus/dbus-connection.c (dbus_connection_dispatch): exit if
2521 exit_on_disconnect flag is set and we process the disconnected
2523 (dbus_connection_set_exit_on_disconnect): new function
2525 2003-09-21 Havoc Pennington <hp@pobox.com>
2527 Get matching rules mostly working in the bus; only actually
2528 parsing the rule text remains. However, the client side of
2529 "signal connections" hasn't been started, this patch is only the
2532 * dbus/dispatch.c: fix for the matching rules changes
2534 * bus/driver.c (bus_driver_handle_remove_match)
2535 (bus_driver_handle_add_match): send an ack reply from these
2538 * glib/dbus-gproxy.c (dbus_gproxy_begin_call): fix order of
2539 arguments, reported by Seth Nickell
2541 * bus/config-parser.c (append_rule_from_element): support
2542 eavesdrop=true|false attribute on policies so match rules
2543 can be prevented from snooping on the system bus.
2545 * bus/dbus-daemon-1.1.in: consistently use terminology "sender"
2546 and "destination" in attribute names; fix some docs bugs;
2547 add eavesdrop=true|false attribute
2549 * bus/driver.c (bus_driver_handle_add_match)
2550 (bus_driver_handle_remove_match): handle AddMatch, RemoveMatch
2553 * dbus/dbus-protocol.h (DBUS_SERVICE_ORG_FREEDESKTOP_BROADCAST): get
2554 rid of broadcast service concept, signals are just always broadcast
2556 * bus/signals.c, bus/dispatch.c, bus/connection.c, bus/bus.c:
2557 mostly implement matching rules stuff (currently only exposed as signal
2560 2003-09-21 Mark McLoughlin <mark@skynet.ie>
2562 * doc/dbus-specification.sgml: Change the header field name
2563 to be an enum and update the rest of the spec to reference
2564 the fields using the conventinal name.
2566 * dbus/dbus-protocol.h: update to reflect the spec.
2568 * doc/TODO: add item to remove the 4 byte alignment requirement.
2570 * dbus/dbus-message.c: Remove the code to generalise the
2571 header/body length and serial number header fields as named
2572 header fields so we can reference field names using the
2574 (append_int_field), (append_uint_field), (append_string_field):
2575 Append the field name as a byte rather than four chars.
2576 (delete_int_or_uint_field), (delete_string_field): reflect the
2577 fact that the field name and typecode now occupy 4 bytes instead
2579 (decode_string_field), (decode_header_data): update to reflect
2580 protocol changes and move the field specific encoding from
2581 decode_string_field() back into decode_header_data().
2583 * dbus/dbus-internals.[ch]: (_dbus_header_field_to_string):
2584 Add utility to aid debugging.
2586 * dbus/dbus-message-builder.c:
2587 (append_string_field), (_dbus_message_data_load): Update to
2588 reflect protocol changes; Change the FIELD_NAME directive
2589 to HEADER_FIELD and allow it to take the field's conventional
2590 name rather than the actual value.
2592 * test/data/*/*.message: Update to use HEADER_FIELD instead
2593 of FIELD_NAME; Always align the header on an 8 byte boundary
2594 *before* updating the header length.
2596 2003-09-15 Havoc Pennington <hp@pobox.com>
2598 * dbus/dbus-pending-call.c: add the get/set object data
2599 boilerplate as for DBusConnection, etc. Use generic object data
2600 for the notify callback.
2602 * glib/dbus-gparser.c (parse_node): parse child nodes
2604 * tools/dbus-viewer.c: more hacking on the dbus-viewer
2606 * glib/dbus-gutils.c (_dbus_gutils_split_path): add a file to
2607 contain functions shared between the convenience lib and the
2610 * glib/Makefile.am (libdbus_glib_1_la_LDFLAGS): add
2611 -export-symbols-regex to the GLib library
2613 * dbus/dbus-object-tree.c (_dbus_object_tree_dispatch_and_unlock):
2614 fix the locking in here, and add a default handler for
2615 Introspect() that just returns sub-nodes.
2617 2003-09-14 Havoc Pennington <hp@pobox.com>
2619 * glib/dbus-gthread.c (dbus_g_thread_init): rename to make g_foo
2620 rather than gfoo consistent
2622 * glib/dbus-gproxy.h: delete for now, move contents to
2623 dbus-glib.h, because the include files don't work right since we
2624 aren't in the dbus/ subdir.
2626 * glib/dbus-gproxy.c (dbus_gproxy_send): finish implementing
2627 (dbus_gproxy_end_call): finish
2628 (dbus_gproxy_begin_call): finish
2630 * glib/dbus-gmain.c (dbus_set_g_error): new
2632 * glib/dbus-gobject.c (handle_introspect): include information
2633 about child nodes in the introspection
2635 * dbus/dbus-connection.c (dbus_connection_list_registered): new
2636 function to help in implementation of introspection
2638 * dbus/dbus-object-tree.c
2639 (_dbus_object_tree_list_registered_and_unlock): new function
2641 2003-09-12 Havoc Pennington <hp@pobox.com>
2643 * glib/dbus-gidl.h: add common base class for all the foo_info
2646 * tools/dbus-viewer.c: add GTK-based introspection UI thingy
2649 * test/Makefile.am: try test srcdir -ef . in addition to test
2650 srcdir = ., one of them should work (yeah lame)
2652 * glib/Makefile.am: build the "idl" parser stuff as a convenience
2655 * glib/dbus-gparser.h: make description_load routines return
2656 NodeInfo* not Parser*
2658 * Makefile.am (SUBDIRS): build test dir after all library dirs
2660 * configure.in: add GTK+ detection
2662 2003-09-07 Havoc Pennington <hp@pobox.com>
2664 * Make Doxygen contented.
2666 2003-09-07 Havoc Pennington <hp@pobox.com>
2668 * doc/dbus-specification.sgml: more updates
2670 2003-09-06 Havoc Pennington <hp@pobox.com>
2672 * doc/dbus-specification.sgml: partial updates
2674 * bus/dbus-daemon-1.1.in: fix the config file docs for the
2675 zillionth time; hopefully I edited the right file this time.
2677 * bus/config-parser.c (append_rule_from_element): support
2678 send_type, send_path, receive_type, receive_path
2680 * bus/policy.c: add message type and path to the list of things
2681 that can be "firewalled"
2683 2003-09-06 Havoc Pennington <hp@pobox.com>
2685 * dbus/dbus-connection.c (dbus_connection_register_fallback): add this
2686 (dbus_connection_register_object_path): make this not handle
2687 messages to paths below the given path
2689 2003-09-03 Havoc Pennington <hp@pobox.com>
2691 * test/glib/Makefile.am: add this with random glib-linked test
2694 * glib/Makefile.am: remove the random test programs from here,
2695 leave only the unit tests
2697 * glib/dbus-gobject.c (_dbus_gobject_test): add test for
2698 uscore/javacaps conversion, and fix
2699 (get_object_property, set_object_property): change to .NET
2700 convention for mapping props to methods, set_FooBar/get_FooBar,
2701 since one language has such a convention we may as well copy it.
2702 Plus real methods in either getFooBar or get_foo_bar style won't
2703 collide with this convention.
2705 2003-09-01 Havoc Pennington <hp@pobox.com>
2707 * glib/dbus-gparser.c: implement
2709 * glib/dbus-gobject.c: start implementing skeletons support
2711 * configure.in: when disabling checks/assert, also define
2712 G_DISABLE_ASSERT and G_DISABLE_CHECKS
2714 2003-09-01 Havoc Pennington <hp@pobox.com>
2716 * glib/Makefile.am: rearrange a bunch of files and get "make
2717 check" framework set up
2719 2003-08-31 Havoc Pennington <hp@pobox.com>
2721 * fix build with --disable-tests
2723 2003-08-30 Havoc Pennington <hp@pobox.com>
2725 * dbus/dbus-connection.c: purge DBusMessageHandler
2727 * dbus/dbus-message-handler.c: remove DBusMessageHandler, just
2728 use callbacks everywhere
2730 2003-08-30 Havoc Pennington <hp@pobox.com>
2732 * test/data/valid-config-files/system.d/test.conf: change to
2733 root for the user so warnings don't get printed
2735 * dbus/dbus-message.c: add dbus_message_get_path,
2736 dbus_message_set_path
2738 * dbus/dbus-object-tree.c (do_test_dispatch): add test of
2739 dispatching to a path
2741 * dbus/dbus-string.c (_dbus_string_validate_path): add
2743 * dbus/dbus-marshal.c (_dbus_demarshal_object_path): implement
2744 (_dbus_marshal_object_path): implement
2746 * dbus/dbus-protocol.h (DBUS_HEADER_FIELD_PATH): new header field
2747 to contain the path to the target object
2748 (DBUS_HEADER_FIELD_SENDER_SERVICE): rename
2749 DBUS_HEADER_FIELD_SENDER to explicitly say it's the sender service
2751 2003-08-30 Havoc Pennington <hp@pobox.com>
2753 * dbus/dbus-object-tree.c: write tests and fix the discovered bugs
2755 2003-08-29 Havoc Pennington <hp@pobox.com>
2757 * dbus/dbus-object-tree.c: modify to allow overlapping paths to be
2759 (struct DBusObjectSubtree): shrink this
2760 a lot, since we may have a lot of them
2761 (_dbus_object_tree_free_all_unlocked): implement
2762 (_dbus_object_tree_dispatch_and_unlock): implement
2764 2003-08-29 Havoc Pennington <hp@pobox.com>
2766 * dbus/dbus-internals.h: fix _DBUS_N_GLOBAL_LOCKS
2768 2003-08-28 Havoc Pennington <hp@pobox.com>
2772 * dbus/dbus-connection.c: port to no ObjectID, create a
2773 DBusObjectTree, rename ObjectTree to ObjectPath in public API
2775 * dbus/dbus-connection.h (struct DBusObjectTreeVTable): delete
2776 everything except UnregisterFunction and MessageFunction
2778 * dbus/dbus-marshal.c: port away from DBusObjectID,
2779 add DBUS_TYPE_OBJECT_PATH
2781 * dbus/dbus-object-registry.[hc], dbus/dbus-object.[hc],
2782 dbus/dbus-objectid.[hc]: remove these, we are moving to
2783 path-based object IDs
2785 2003-08-25 Havoc Pennington <hp@pobox.com>
2787 Just noticed that dbus_message_test is hosed, I wonder when I
2788 broke that. I thought make check was passing earlier...
2790 * dbus/dbus-object-tree.c: add new "object tree" to match DCOP
2791 container tree, will replace most of dbus-object-registry
2793 * dbus/dbus-string.c (_dbus_string_append_printf_valist): fix C99
2796 2003-08-19 Havoc Pennington <hp@pobox.com>
2798 * dbus/dbus-message.c (decode_string_field): support FIELD_SENDER
2799 (dbus_message_is_error): fix this function
2801 * bus/dbus-daemon-1.1: clarify logic on when <deny>/<allow> rules
2804 * bus/policy.c (bus_client_policy_check_can_receive): fix code to
2805 reflect clarified man page
2806 (bus_client_policy_check_can_send): ditto
2808 * bus/session.conf.in: fixup
2810 * bus/system.conf.in: fixup
2812 2003-08-18 Havoc Pennington <hp@redhat.com>
2814 * dbus/dbus-hash.c (_dbus_hash_table_insert_two_strings): fix
2816 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
2817 dumb bug created earlier (wrong order of args to
2818 decode_header_data())
2820 * tools/dbus-send.c: port
2822 * tools/dbus-print-message.c (print_message): port
2824 * test/data/*messages: port all messages over
2826 * dbus/dbus-message-builder.c: support including
2829 * bus/driver.c: port over
2831 * bus/dispatch.c: port over to new stuff
2833 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
2834 rename disconnect signal to "Disconnected"
2836 2003-08-17 Havoc Pennington <hp@pobox.com>
2838 This doesn't compile yet, but syncing up so I can hack on it from
2839 work. What are branches for if not broken code? ;-)
2841 * dbus/dbus-protocol.h: remove DBUS_HEADER_FIELD_NAME, add
2842 DBUS_HEADER_FIELD_INTERFACE, DBUS_HEADER_FIELD_MEMBER,
2843 DBUS_HEADER_FIELD_ERROR_NAME
2845 * dbus/dbus-hash.c: Introduce DBUS_HASH_TWO_STRINGS as hack to use
2846 for the interface+member pairs
2847 (string_hash): change to use g_str_hash algorithm
2848 (find_direct_function, find_string_function): refactor these to
2851 * dbus/dbus-message.c: port all of this over to support
2852 interface/member fields instead of name field
2854 * dbus/dbus-object-registry.c: port over
2856 * dbus/dbus-string.c (_dbus_string_validate_interface): rename
2857 from _dbus_string_validate_name
2859 * bus/dbus-daemon-1.1: change file format for the
2860 <deny>/<allow> stuff to match new message naming scheme
2862 * bus/policy.c: port over
2864 * bus/config-parser.c: parse new format
2866 2003-08-16 Havoc Pennington <hp@pobox.com>
2868 * dbus/dbus-object-registry.c (add_and_remove_objects): remove
2871 * glib/dbus-gproxy.c: some hacking
2873 2003-08-15 Havoc Pennington <hp@redhat.com>
2875 * dbus/dbus-pending-call.c (dbus_pending_call_block): implement
2877 * dbus/dbus-connection.c
2878 (dbus_connection_send_with_reply_and_block): factor out internals;
2879 change to convert any error replies to DBusError instead of
2880 returning them as a message
2882 2003-08-15 Havoc Pennington <hp@pobox.com>
2884 * dbus/dbus-connection.c,
2885 dbus/dbus-pending-call.c: Finish the pending call stuff
2887 2003-08-14 Havoc Pennington <hp@redhat.com>
2889 * dbus/dbus-pending-call.c: start on new object that will replace
2890 DBusMessageHandler and ReplyHandlerData for tracking outstanding
2893 * dbus/dbus-gproxy.c: start on proxy object used to communicate
2894 with remote interfaces
2896 * dbus/dbus-gidl.c: do the boring boilerplate in here
2898 2003-08-12 Havoc Pennington <hp@pobox.com>
2900 * bus/dispatch.c (bus_dispatch): make this return proper
2901 DBusHandlerResult to avoid DBUS_ERROR_UNKNOWN_METHOD
2903 * dbus/dbus-errors.c (dbus_set_error): use
2904 _dbus_string_append_printf_valist
2906 * dbus/dbus-string.c (_dbus_string_append_printf_valist)
2907 (_dbus_string_append_printf): new
2909 * dbus/dbus-errors.h (DBUS_ERROR_UNKNOWN_MESSAGE): change to
2912 * dbus/dbus-connection.c (dbus_connection_dispatch): handle
2913 DBUS_HANDLER_RESULT_NEED_MEMORY; send default error reply if a
2914 message is unhandled.
2916 2003-08-11 Havoc Pennington <hp@pobox.com>
2918 * bus/test.c (client_disconnect_handler): change to return
2919 HANDLED (would have been REMOVE_MESSAGE)
2921 * dbus/dbus-object.h (enum DBusHandlerResult): rename to
2922 HANDLED/NOT_YET_HANDLED instead of
2923 REMOVE_MESSAGE/ALLOW_MORE_HANDLERS to make it clearer how it
2926 2003-08-10 Havoc Pennington <hp@pobox.com>
2928 * tools/dbus-send.c (main): add --type argument, for now
2929 supporting only method_call and signal types.
2931 * tools/dbus-print-message.c: print message type
2933 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
2934 init connection->objects
2936 * doc/dbus-specification.sgml: fix sgml
2938 * bus/*.c: port over to object-instance API changes
2940 * test/test-service.c: ditto
2942 * dbus/dbus-message.c (dbus_message_create_header): allow #NULL
2943 name, we will have to fix up the rest of the code to also handle
2945 (dbus_message_new): generic message-creation call
2946 (set_string_field): allow appending name field
2948 2003-08-06 Havoc Pennington <hp@pobox.com>
2950 * dbus/dbus-object-registry.c: implement signal connection
2953 * dbus/dbus-connection.c (_dbus_connection_unref_unlocked): new
2955 * dbus/dbus-internals.c (_dbus_memdup): new function
2957 2003-08-02 Havoc Pennington <hp@pobox.com>
2959 * dbus/dbus-message.c (dbus_message_get_no_reply)
2960 (dbus_message_set_no_reply): add these and remove
2961 set_is_error/get_is_error
2963 * dbus/dbus-protocol.h, doc/dbus-specification.sgml:
2964 remove the ERROR flag, since there's now an ERROR type
2966 2003-08-01 Havoc Pennington <hp@pobox.com>
2968 * dbus/dbus-object-registry.c (_dbus_object_registry_handle_and_unlock):
2971 * dbus/dbus-message.c (dbus_message_get_type): new function
2973 * doc/dbus-specification.sgml: add "type" byte to messages
2975 2003-08-01 Havoc Pennington <hp@pobox.com>
2977 * dbus/dbus-protocol.h (DBUS_MESSAGE_TYPE_*): introduce
2978 a message type enum to distinguish kinds of message
2979 (DBUS_HEADER_FLAG_NO_REPLY_EXPECTED): flag for a message
2980 that need not be replied to
2982 2003-08-01 Havoc Pennington <hp@pobox.com>
2984 * dbus/dbus-marshal.c: adapt to DBusObjectID changes
2985 (unpack_8_octets): fix no-64-bit-int bug
2987 * dbus/dbus-object-registry.c (validate_id): validate the
2988 connection ID bits, not just the instance ID.
2990 * dbus/dbus-connection.c (_dbus_connection_init_id): initialize
2991 the connection-global 33 bits of the object ID
2993 * dbus/dbus-object-registry.c (info_from_entry): fill in
2994 object ID in the new way
2996 * dbus/dbus-objectid.h: rather than high/low bits, specifically
2997 define server/client/instance bits.
2999 2003-07-30 Havoc Pennington <hp@pobox.com>
3001 * dbus/dbus-connection.c (dbus_connection_register_object): fix
3004 2003-07-13 Havoc Pennington <hp@pobox.com>
3006 * dbus/dbus-object.h (struct DBusObjectVTable): add padding
3007 fields to DBusObjectVTable and DBusObjectInfo
3009 2003-07-12 Havoc Pennington <hp@pobox.com>
3011 * dbus/dbus-object-registry.c: implement unit test,
3012 fix bugs discovered in process
3014 * dbus/dbus-connection.c: remove handler_table and
3015 register_handler(), add DBusObjectRegistry usage
3017 * dbus/dbus-objectid.c (dbus_object_id_is_null)
3018 (dbus_object_id_set_null): new functions
3020 2003-07-08 Havoc Pennington <hp@pobox.com>
3022 * dbus/dbus-object.c: implement some of this
3024 * dbus/dbus-object-registry.c
3025 (_dbus_object_registry_add_and_unlock): fill in the object_id out
3027 (_dbus_object_registry_new): handle OOM
3029 2003-07-08 Havoc Pennington <hp@pobox.com>
3031 * dbus/dbus-object.h: sketch out an API for registering objects
3032 with a connection, that allows us to use as little as 24 bytes
3033 per object and lets application code represent an object in
3034 any conceivable way.
3036 * dbus/dbus-object-registry.c: implement the hard bits of the
3037 DBusConnection aspect of object API. Not yet wired up.
3039 2003-07-06 Havoc Pennington <hp@pobox.com>
3041 * dbus/dbus-marshal.c (_dbus_marshal_set_object_id): new function
3042 (_dbus_marshal_object_id): new
3043 (_dbus_demarshal_object_id): new
3044 (_dbus_marshal_get_arg_end_pos): support object ID type, and
3045 consolidate identical switch cases. Don't conditionalize handling
3046 of DBUS_TYPE_UINT64, need to handle the type always.
3047 (_dbus_marshal_validate_arg): consolidate identical cases, and
3048 handle DBUS_TYPE_OBJECT_ID
3050 * dbus/dbus-objectid.c: new file with DBusObjectID data type.
3052 * dbus/dbus-protocol.h: add DBUS_TYPE_OBJECT_ID
3054 2003-09-28 Havoc Pennington <hp@pobox.com>
3058 2003-09-28 Havoc Pennington <hp@pobox.com>
3060 * doc/Makefile.am (dbus-specification.html): testing a funky hack
3061 to work with Debian db2html
3063 2003-09-28 Havoc Pennington <hp@pobox.com>
3065 * configure.in: 0.13
3067 * doc/Makefile.am (dbus-test-plan.html): accept nonexistence of
3068 stylesheet-images for benefit of Debian
3070 Change back to using filesystem-linked sockets for the system
3071 bus, so only root can create the default system bus address.
3073 * bus/system.conf.in: change to use
3074 DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
3076 * dbus/Makefile.am (INCLUDES): remove DBUS_SYSTEM_BUS_PATH define
3079 * configure.in: define DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
3080 here, and AC_DEFINE DBUS_SYSTEM_PATH
3082 2003-08-09 Anders Carlsson <andersca@codefactory.se>
3085 * doc/busconfig.dtd:
3088 2003-08-09 Anders Carlsson <andersca@codefactory.se>
3090 * doc/dbus-specification.sgml:
3091 Add activation reply values.
3093 2003-08-05 Havoc Pennington <hp@redhat.com>
3095 * configure.in: 0.12
3097 2003-08-05 Anders Carlsson <andersca@codefactory.se>
3099 * glib/dbus-gmain.c: (watch_fd_new), (watch_fd_ref),
3100 (watch_fd_unref), (dbus_gsource_check), (dbus_gsource_dispatch),
3101 (add_watch), (remove_watch), (create_source):
3102 Refcount fds, fixes some reentrancy issues.
3104 2003-07-30 Havoc Pennington <hp@redhat.com>
3106 * dbus/dbus-bus.c (init_connections_unlocked): fix default system
3107 bus address to be abstract if we have abstract sockets
3111 2003-07-28 Havoc Pennington <hp@redhat.com>
3113 * bus/messagebus.in: fix to avoid processname/servicename
3114 confusion, from Michael Kearey
3115 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=100965
3117 2003-07-23 Havoc Pennington <hp@pobox.com>
3119 * dbus/dbus-message.c (dbus_message_iter_get_named):
3120 fix from Andy Hanton to remove broken "+1"
3122 2003-07-16 Havoc Pennington <hp@pobox.com>
3124 * tools/dbus-launch.c (babysit): close stdout/stderr in the
3125 babysitter process, as suggested by Thomas Leonard, so
3126 an "eval `dbus-launch --exit-with-session`" will actually
3129 2003-07-16 Havoc Pennington <hp@pobox.com>
3131 * configure.in: print out EXPANDED_* variables in the summary at
3132 the end; clean up the code that computes EXPANDED_ variables and
3133 get the ones using exec_prefix right. Should make things work
3134 when you build without --prefix
3136 2003-06-29 Havoc Pennington <hp@pobox.com>
3138 * mono/Test.cs (class Test): fire up a main loop and run it
3140 * mono/DBus.cs (DBus): don't g_thread_init since it can only be
3141 done once, the app has to do it
3143 2003-06-26 Havoc Pennington <hp@pobox.com>
3145 * mono/Connection.cs: set up connection with the glib main loop
3147 2003-07-01 Havoc Pennington <hp@redhat.com>
3149 * doc/dbus-specification.sgml: clarify the format of a type code,
3150 change suggested by Jim Blandy
3152 2003-06-29 Miloslav Trmac <mitr@volny.cz>
3155 * tools/Makefile.am: Don't assume srcdir == builddir.
3157 * dbus/dbus-memory.c (dbus_realloc): Don't check guards after shrinking
3158 the allocated block.
3159 (_dbus_memory_test): New function.
3160 * dbus/dbus-test.h: Add _dbus_memory_test ().
3161 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): Call it.
3163 * dbus/dbus-message.c (decode_header_data): Use %.4s instead
3165 (dbus_message_new): Remove obsolete @todo.
3167 * dbus/dbus-marshal.c (_dbus_marshal_set_int64)
3168 (_dbus_marshal_set_uint64): Fix comment.
3170 * dbus/dbus-message.c (append_int_field, append_uint_field): Don't
3171 hardcode FIELD_REPLY_SERIAL.
3173 * dbus/dbus-mainloop.c (_dbus_loop_remove_watch)
3174 (_dbus_loop_remove_timeout): Cast function pointers to (void *) for %p
3176 * configure.in: Add -D_POSIX_C_SOURCE=199309L -DBSD_SOURCE to CFLAGS
3177 and disable DBUS_USE_ATOMIC_INT_486 when --enable-ansi is used
3179 2003-06-24 Havoc Pennington <hp@pobox.com>
3181 * mono/*.cs: Use IntPtr.Zero instead of ((IntPtr) 0)
3183 2003-06-23 Anders Carlsson <andersca@codefactory.se>
3189 * gcj/TestMessage.java: (TestMessage), (TestMessage.main):
3190 * gcj/org/.cvsignore:
3191 * gcj/org/Makefile.am:
3192 * gcj/org/freedesktop/.cvsignore:
3193 * gcj/org/freedesktop/Makefile.am:
3194 * gcj/org/freedesktop/dbus/.cvsignore:
3195 * gcj/org/freedesktop/dbus/Makefile.am:
3196 * gcj/org/freedesktop/dbus/Message.java: (Message),
3198 * gcj/org/freedesktop/dbus/natMessage.cc:
3199 Fix the build system.
3201 2003-06-22 Havoc Pennington <hp@pobox.com>
3203 * mono/Connection.cs: add more bindings
3205 * dbus/dbus-threads.c (dbus_threads_init): allow calling this
3208 2003-06-22 Havoc Pennington <hp@pobox.com>
3210 * mono/Connection.cs, mono/DBus.cs, mono/Error.cs:
3211 Start wrapping more stuff.
3213 2003-06-22 Havoc Pennington <hp@pobox.com>
3215 * mono/Message.cs: implement Message.Wrap() that ensures we only
3216 have a single C# wrapper per DBusMessage, assuming it works which
3217 it probably doesn't.
3219 * dbus/dbus-message.c (dbus_message_allocate_data_slot): new
3220 (dbus_message_free_data_slot): new
3221 (dbus_message_set_data): new
3222 (dbus_message_get_data): new
3224 2003-06-22 Havoc Pennington <hp@pobox.com>
3226 * dbus/dbus-dataslot.c (_dbus_data_slot_allocator_unref)
3227 (_dbus_data_slot_allocator_alloc): rework these to keep a
3228 reference count on each slot and automatically manage a global
3229 slot ID variable passed in by address
3231 * bus/bus.c: convert to new dataslot API
3233 * dbus/dbus-bus.c: convert to new dataslot API
3235 * dbus/dbus-connection.c: convert to new dataslot API
3237 * dbus/dbus-server.c: convert to new dataslot API
3239 * glib/dbus-gmain.c: ditto
3243 * bus/connection.c: ditto
3245 2003-06-22 Anders Carlsson <andersca@codefactory.se>
3247 * configure.in: Add AM_PROG_GCJ and move AM_PROG_LIBTOOL
3248 after the gcj checks so that the correct configuration tags
3249 will be added to libtool.
3251 * dbus-glib-1.pc.in: No need to specify any includes since
3252 dbus-1.pc.in has those.
3254 2003-06-22 Havoc Pennington <hp@pobox.com>
3256 * mono/*, gcj/*, configure.in, Makefile.am:
3257 Check in makefiles and subdirs for mono and gcj bindings.
3258 Neither binding actually exists, just trying to get through
3259 all the build and other boring bits.
3261 2003-06-21 Philip Blundell <philb@gnu.org>
3263 * tools/dbus-monitor.1: Updated.
3265 * tools/dbus-send.1: Likewise.
3267 2003-06-20 Anders Carlsson <andersca@codefactory.se>
3269 * dbus/dbus-transport-unix.c (unix_handle_watch): Check
3270 for hangup and error after checking read so we won't discard
3271 pending data if both hangup and read are set.
3273 2003-06-19 Philip Blundell <philb@gnu.org>
3275 * tools/dbus-print-message.c (print_message): Handle BOOLEAN.
3277 * tools/dbus-send.c: Accept both --system and --session.
3279 * tools/dbus-monitor.c: Same here.
3281 2003-06-19 Anders Carlsson <andersca@codefactory.se>
3283 * glib/dbus-glib.h: Fix so that dbus-glib.h can be used
3284 from C++ (Patch by Miloslav Trmac).
3286 2003-06-15 Joe Shaw <joe@assbarn.com>
3288 * configure.in: Check for socklen_t.
3290 * dbus/dbus-sysdeps.c: Define socklen_t if it's not defined.
3292 * test/test-segfault.c: Add #include <sys/time.h>
3294 * tools/Makefile.am: Add DBUS_X_CFLAGS to the INCLUDES since
3295 dbus-launch needs it.
3297 2003-06-09 Havoc Pennington <hp@redhat.com>
3299 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): don't use
3300 SUN_LEN, it breaks abstract socket usage
3302 * dbus/dbus-internals.c (_dbus_verbose_real): only print PID at
3305 2003-06-04 Havoc Pennington <hp@pobox.com>
3307 * dbus/dbus-server.c (dbus_server_listen): allow abstract sockets
3308 using unix:abstract=/foo, and when listening in a tmpdir
3309 i.e. unix:tmpdir=/tmp, always use abstract sockets if we can.
3311 * dbus/dbus-transport.c (_dbus_transport_open): support
3314 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
3315 support abstract sockets
3317 * dbus/dbus-transport-unix.c
3318 (_dbus_transport_new_for_domain_socket): support abstract sockets
3320 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket): add "abstract"
3321 toggle as an argument, implement abstract namespace support
3322 (_dbus_listen_unix_socket): ditto
3324 * configure.in: add --enable-abstract-sockets and implement
3325 a configure check for autodetection of the right value.
3327 2003-06-01 Havoc Pennington <hp@pobox.com>
3329 * tools/dbus-cleanup-sockets.c: add utility to clean up sockets
3330 in /tmp (though on Linux this will end up being useless,
3331 when we add abstract namespace support)
3333 * configure.in: define DBUS_SESSION_SOCKET_DIR in addition to
3336 2003-05-28 Colin Walters <walters@verbum.org>
3338 * tools/dbus-monitor.c (main): Fix silly typo (s/--session/--system/).
3340 2003-05-18 Anders Carlsson <andersca@codefactory.se>
3342 * dbus/dbus-message.c (dbus_message_new): Remove @todo.
3344 2003-05-17 Colin Walters <walters@gnu.org>
3346 * tools/dbus-send.c: Don't exit with an error code if --help was
3347 passed. Default to using the session bus instead of the system
3350 * tools/dbus-launch.c: Ditto.
3352 * tools/dbus-monitor.c: Ditto.
3354 * tools/dbus-send.1: Update with new arguments.
3356 * tools/dbus-launch.c: Emit code to export variables. New
3357 arguments -s and -c to specify shell syntax, and a bit of code to
3358 autodetect syntax. Also, allow specifying a program to run.
3360 * tools/dbus-launch.1: Update with new arguments.
3362 * tools/dbus-send.1: Ditto.
3364 * tools/dbus-monitor.1: Ditto.
3366 2003-05-17 Havoc Pennington <hp@pobox.com>
3368 * bus/config-parser.c (merge_included): merge in policies from
3369 child configuration file.
3371 * bus/policy.c (bus_policy_merge): function to merge two policies
3374 2003-05-16 Havoc Pennington <hp@redhat.com>
3376 * dbus/dbus-connection.c: disable verbose lock spew
3378 * tools/dbus-send.c: add --print-reply command line option
3380 * tools/dbus-print-message.h (print_message): new util function
3381 shared by dbus-send and dbus-monitor
3383 * tools/dbus-monitor.c (handler_func): exit on disconnect
3385 * dbus/dbus-transport-unix.c (do_reading): if the transport is
3386 disconnected, don't try to use the read_watch
3388 * dbus/dbus-watch.c (dbus_watch_get_enabled): assert watch != NULL
3389 so we can find this bug more easily
3391 2003-05-16 Havoc Pennington <hp@redhat.com>
3393 * bus/policy.c (free_rule_list_func): avoid a crash when passed
3394 NULL as DBusHashTable is annoyingly likely to do.
3396 2003-05-16 Colin Walters <walters@verbum.org>
3398 * tools/dbus-monitor.c: Add --session argument and usage()
3401 * tools/dbus-monitor.1: Update with new --session arg.
3403 * bus/Makefile.am (install-data-hook): Create
3404 $(libdir)/dbus-1.0/services so that the session bus is happy.
3406 2003-05-15 Havoc Pennington <hp@redhat.com>
3408 * dbus/dbus-sysdeps.c (_dbus_atomic_dec, _dbus_atomic_inc): work
3409 on non-x86. ifdef's are evil.
3411 2003-05-15 Havoc Pennington <hp@redhat.com>
3413 * configure.in: 0.11
3417 * bus/Makefile.am (initddir): apparently we are supposed to put
3418 init scripts in /etc/rc.d/init.d not /etc/init.d
3420 * bus/Makefile.am: remove the "you must --enable-tests to make
3421 check" as it broke distcheck
3423 * bus/Makefile.am (install-data-hook): create /etc/dbus-1/system.d
3425 2003-05-13 James Willcox <jwillcox@gnome.org>
3428 * bus/activation.c: (bus_activation_service_created),
3429 (bus_activation_activate_service):
3430 * bus/driver.c: (bus_driver_send_service_deleted),
3431 (bus_driver_send_service_created), (bus_driver_send_service_lost),
3432 (bus_driver_send_service_acquired),
3433 (bus_driver_send_welcome_message),
3434 (bus_driver_handle_list_services):
3435 * bus/session.conf.in:
3436 * dbus/dbus-bus.c: (dbus_bus_acquire_service),
3437 (dbus_bus_service_exists), (dbus_bus_activate_service):
3440 Add some convenience API which lets you activate a service, and did a
3441 bunch of s/0/DBUS_TYPE_INVALID/ in calls to dbus_message_append_args()
3442 and dbus_message_get_args()
3444 2003-05-11 Havoc Pennington <hp@pobox.com>
3446 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix to avoid
3447 calling _dbus_marshal_validate_arg() for every byte in a byte
3450 * dbus/dbus-message-handler.c: use atomic reference counting to
3451 reduce number of locks slightly; the global lock in here sucks
3453 * dbus/dbus-connection.c
3454 (_dbus_connection_update_dispatch_status_and_unlock): variant of
3455 update_dispatch_status that can be called with lock held; then use
3456 in a couple places to reduce locking/unlocking
3457 (dbus_connection_send): hold the lock over the whole function
3458 instead of acquiring it twice.
3460 * dbus/dbus-timeout.c (_dbus_timeout_new): handle OOM
3462 * bus/connection.c (bus_connections_setup_connection): fix access
3463 to already-freed memory.
3465 * dbus/dbus-connection.c: keep a little cache of linked list
3466 nodes, to avoid using the global linked list alloc lock in the
3467 normal send-message case. Instead we just use the connection lock
3468 that we already have to take.
3470 * dbus/dbus-list.c (_dbus_list_find_last): new function
3472 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec):
3473 change to use a struct for the atomic type; fix docs,
3474 they return value before increment, not after increment.
3476 * dbus/dbus-string.c (_dbus_string_append_4_aligned)
3477 (_dbus_string_append_8_aligned): new functions to try to
3478 microoptimize this operation.
3479 (reallocate_for_length): break this out of set_length(), to
3480 improve profile info, and also so we can consider inlining the
3483 * dbus/dbus-message.c (dbus_message_new_empty_header): init data
3484 strings with some preallocation, cuts down on our calls to realloc
3485 a fair bit. Though if we can get the "move entire string to empty
3486 string" optimization below to kick in here, it would be better.
3488 * dbus/dbus-string.c (_dbus_string_move): just call
3489 _dbus_string_move_len
3490 (_dbus_string_move_len): add a special case for moving
3491 an entire string into an empty string; we can just
3492 swap the string data instead of doing any reallocs.
3493 (_dbus_string_init_preallocated): new function
3495 2003-05-11 Havoc Pennington <hp@pobox.com>
3497 Write a "test-profile" that does echo client-server with threads;
3498 profile reveals lock contention, memcpy/realloc of buffers, and
3499 UTF-8 validation as hot spots. 20% of lock contention eliminated
3500 with dbus_atomic_inc/dec implementation on x86. Much remaining
3501 contention is global mempool locks for GList and DBusList.
3503 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec): add
3506 * dbus/dbus-connection.c (struct DBusConnection): use
3507 dbus_atomic_t for the reference count
3509 * dbus/dbus-message.c (struct DBusMessage): declare
3510 dbus_atomic_t values as volatile
3512 * configure.in: code to detect ability to use atomic integer
3513 operations in assembly, from GLib patch
3515 * dbus/dbus-internals.c (_dbus_verbose_real): call getpid every
3516 time, tired of it being wrong in threads and forked processes
3518 * glib/test-profile.c: a little program to bounce messages back
3519 and forth between threads and eat CPU
3521 * dbus/dbus-connection.c: add debug spew macros for debugging
3522 thread locks; include config.h at top; fix deadlock in
3523 dbus_connection_flush()
3525 2003-05-08 Havoc Pennington <hp@pobox.com>
3527 * dbus/dbus-spawn.c: s/_exit/exit/ because it was keeping gcov
3528 data from getting written, and there wasn't a good reason to
3531 * test/decode-gcov.c (mark_inside_dbus_build_tests): don't count
3532 dbus_verbose lines in test coverage
3533 (main): add list of functions sorted by # of untested blocks
3534 to the coverage report
3536 * dbus/dbus-mempool.c: put some test-only code in DBUS_BUILD_TESTS
3538 * dbus/dbus-marshal.c (_dbus_marshal_test): extend test coverage
3540 * dbus/dbus-message-handler.c (_dbus_message_handler_test):
3541 extend test coverage
3543 * test/data/auth/cancel.auth-script: test canceling an
3546 * dbus/Makefile.am: remove dbus-server-debug.[hc] for now, as they
3547 aren't used. in CVS history if we end up needing them.
3549 2003-05-04 Havoc Pennington <hp@pobox.com>
3551 * dbus/dbus-message-handler.c (_dbus_message_handler_test): add
3554 * dbus/dbus-marshal.c (_dbus_demarshal_string_array): fix this
3555 function, which assumed length was in # of strings, not bytes
3557 * dbus/dbus-message.c (_dbus_message_test): add tests for some
3560 * dbus/dbus-connection.c
3561 (_dbus_connection_queue_received_message): disable function for
3562 now, we are only using it in test mode
3564 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
3565 remove a mistaken FIXME
3567 2003-05-04 Havoc Pennington <hp@pobox.com>
3569 * dbus/dbus-connection.c (dbus_connection_preallocate_send):
3570 unlock mutex on successful return, patch from Anders Gustafsson
3572 2003-05-04 Havoc Pennington <hp@pobox.com>
3574 * dbus-glib-1.pc.in (Requires): fix dependencies, from
3577 2003-05-04 Havoc Pennington <hp@pobox.com>
3579 * tools/dbus-launch.c: implement
3581 * bus/main.c (main), bus/bus.c (bus_context_new):
3582 implement --print-pid and --fork
3584 2003-05-03 Havoc Pennington <hp@redhat.com>
3586 * dbus/dbus-address.c (dbus_parse_address): fix bug when a key in
3587 the address had no value, and add to test suite. Fix and
3588 regression test from Miloslav Trmac
3590 2003-05-03 Havoc Pennington <hp@pobox.com>
3592 * dbus/dbus-watch.c (dbus_watch_handle): warn and return if a
3593 watch is invalid when handled
3595 * tools/Makefile.am, tools/dbus-launch.c, tools/dbus-launch.1: add
3596 dbus-launch utility to launch the bus from a shell script. Didn't
3597 actually implement dbus-launch yet, it's just a placeholder still.
3599 2003-05-03 Havoc Pennington <hp@pobox.com>
3601 * bus/Makefile.am, bus/dbus-daemon-1.1.in: man page for the
3602 daemon; also documents daemon config file, so replaces
3603 doc/config-file.txt. Corrected some stuff from config-file.txt in
3604 the process of moving it.
3606 2003-05-03 Havoc Pennington <hp@pobox.com>
3608 * tools/Makefile.am, tools/dbus-send.1, tools/dbus-monitor.1:
3611 2003-05-03 Colin Walters <walters@verbum.org>
3613 * dbus/dbus-sysdeps.c (fill_user_info): Test against
3614 DBUS_UID_UNSET to determine whether to do a uid lookup or not.
3616 * Makefile.am: Update to use new .pc versioning scheme.
3618 2003-05-02 Havoc Pennington <hp@redhat.com>
3620 * bus/system.conf.in: allow send/receive to/from message bus
3623 2003-04-30 Havoc Pennington <hp@redhat.com>
3625 * configure.in: print a note when building with unit tests and
3628 2003-04-30 Havoc Pennington <hp@redhat.com>
3630 * Makefile.am: add a check-local that complains if you didn't
3631 configure with --enable-tests
3633 2003-04-29 Havoc Pennington <hp@redhat.com>
3635 * glib/dbus-gmain.c: docs cleanups
3637 * dbus/dbus-types.h: add docs on int64 types
3639 * dbus/dbus-memory.c: fix docs to avoid putting private API in
3640 public API docs section
3642 2003-04-29 Havoc Pennington <hp@redhat.com>
3644 * dbus-1.pc.in, dbus-glib-1.pc.in: rename these from
3645 dbus-1.0.pc.in, dbus-glib-1.0.pc.in. As these change with the
3646 parallel install API version, not with the D-BUS package version.
3648 * HACKING: move some of README over here
3650 * README: updates, and document API/ABI policy
3652 * configure.in: reindentation
3654 2003-04-29 Havoc Pennington <hp@redhat.com>
3656 * dbus/dbus.h: add "you have to define DBUS_API_SUBJECT_TO_CHANGE
3657 to use this library" to be sure people have the right
3660 2003-04-28 Havoc Pennington <hp@redhat.com>
3662 * configure.in: add --enable-docs which by default is auto yes if
3663 doxygen and db2html found, no otherwise; but can be forced on/off
3665 * doc/Makefile.am: conditionalize whether to build docs on
3668 2003-04-28 Havoc Pennington <hp@redhat.com>
3670 * configure.in: 0.10
3674 * bus/system.conf.in: add <includedir>system.d</includedir>
3676 * dbus/dbus-userdb.c (_dbus_user_database_lookup): fix bug when
3677 username was provided but not uid
3679 * bus/config-parser.c (struct BusConfigParser): keep track of
3680 whether the parser is toplevel or was included; change some
3681 of the error handling if it's included.
3683 2003-04-27 Havoc Pennington <hp@pobox.com>
3687 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
3688 report correct status if we finish processing authentication
3689 inside this function.
3691 * bus/activation.c (try_send_activation_failure): use
3692 bus_transaction_send_error_reply
3694 * bus/connection.c (bus_connection_get_groups): return an error
3695 explaining the problem
3697 * bus/bus.c (bus_context_check_security_policy): implement
3698 restriction here that inactive connections can only send the
3699 hello message. Also, allow bus driver to send anything to
3702 * bus/connection.c (bus_connection_complete): create the
3703 BusClientPolicy here instead of on-demand.
3704 (bus_connection_get_policy): don't return an error
3706 * dbus/dbus-message.c (dbus_message_new_error_reply): allow NULL
3707 sender field in message being replied to
3709 * bus/bus.c (bus_context_check_security_policy): fix silly typo
3710 causing it to return FALSE always
3712 * bus/policy.c (bus_client_policy_check_can_send): fix bug where
3713 we checked sender rather than destination
3715 2003-04-25 Havoc Pennington <hp@redhat.com>
3717 test suite is slightly hosed at the moment, will fix soon
3719 * bus/connection.c (bus_connections_expire_incomplete): fix to
3720 properly disable the timeout when required
3721 (bus_connection_set_name): check whether we can remove incomplete
3722 connections timeout after we complete each connection.
3724 * dbus/dbus-mainloop.c (check_timeout): fix this up a bit,
3725 probably still broken.
3727 * bus/services.c (bus_registry_acquire_service): implement max
3728 number of services owned, and honor allow/deny rules on which
3729 services a connection can own.
3731 * bus/connection.c (bus_connection_get_policy): report errors here
3733 * bus/activation.c: implement limit on number of pending
3736 2003-04-25 Havoc Pennington <hp@redhat.com>
3738 * dbus/dbus-transport.c (_dbus_transport_get_unix_user): fix bug
3739 where we used >= 0 instead of != DBUS_UID_UNSET.
3741 2003-04-25 Havoc Pennington <hp@redhat.com>
3743 * glib/dbus-gmain.c (remove_watch): fix for a crash when watches
3744 were toggled without add/remove, fix from Anders Gustafsson
3746 2003-04-24 Havoc Pennington <hp@redhat.com>
3748 * test/data/valid-config-files/basic.conf: add <limit> tags to
3751 * bus/config-parser.h, bus/config-parser.c, bus/bus.c: Implement
3752 <limit> tag in configuration file.
3754 2003-04-24 Havoc Pennington <hp@redhat.com>
3756 * bus/dispatch.c: somehow missed some name_is
3758 * dbus/dbus-timeout.c (_dbus_timeout_set_enabled)
3759 (_dbus_timeout_set_interval): new
3761 * bus/connection.c (bus_connections_setup_connection): record time
3762 when each connection is first set up, and expire them after the
3763 auth timeout passes.
3765 2003-04-24 Havoc Pennington <hp@redhat.com>
3767 * dbus/dbus-message.c (dbus_message_name_is): rename
3768 (dbus_message_service_is): rename
3769 (dbus_message_sender_is): rename
3770 (dbus_message_get_service): rename
3772 2003-04-24 Havoc Pennington <hp@redhat.com>
3774 * configure.in: add --enable-checks
3776 * dbus/dbus-message.c (dbus_message_new): reverse name/service arguments
3778 * dbus/dbus-connection.c (dbus_connection_preallocate_send): fix
3779 to use thread locks.
3780 (_dbus_connection_handler_destroyed_locked): move some private
3781 functions into proper docs group
3783 * dbus/dbus-internals.h: add _dbus_return_if_fail,
3784 _dbus_return_val_if_fail
3786 Throughout: use dbus_return_if_fail
3788 2003-04-23 James Willcox <jwillcox@gnome.org>
3791 * glib/dbus-gmain.c: (add_timeout), (wakeup_main), (create_source),
3792 (dbus_connection_setup_with_g_main),
3793 (dbus_server_setup_with_g_main):
3794 * glib/test-dbus-glib.c: (main):
3795 * glib/test-thread-client.c: (main):
3796 * glib/test-thread-server.c: (new_connection_callback), (main):
3797 * tools/dbus-monitor.c: (main):
3799 Added a GMainContext argument to dbus_connection_setup_with_g_main()
3800 and dbus_server_setup_with_g_main().
3802 2003-04-20 Havoc Pennington <hp@pobox.com>
3804 * doc/dbus-specification.sgml: document the restrictions on
3805 message and service names
3807 2003-04-22 Havoc Pennington <hp@redhat.com>
3809 * dbus/dbus-message.c, dbus/dbus-marshal.c: add 64-bit integer
3810 support, and do some code cleanups to share more code and
3811 speed up array marshal/demarshal.
3813 * dbus-1.0.pc.in (Cflags): put libdir include file in cflags
3815 * configure.in: generate dbus-arch-deps.h
3817 * dbus/dbus-protocol.h (DBUS_TYPE_INT64, DBUS_TYPE_UINT64): add
3820 2003-04-22 Havoc Pennington <hp@redhat.com>
3822 * test/data/valid-messages/opposite-endian.message: fix test
3823 to use proper type for rply field
3825 * test/data/invalid-messages: add tests for below validation
3827 * dbus/dbus-message.c (decode_header_data): validate field types,
3828 and validate that named fields are valid names
3829 (decode_name_field): consider messages in the
3830 org.freedesktop.Local. namespace to be invalid.
3832 * dbus/dbus-string.c (_dbus_string_validate_name): new
3834 2003-04-19 Havoc Pennington <hp@pobox.com>
3836 * bus/driver.c (bus_driver_handle_hello): check limits and
3837 return an error if they are exceeded.
3839 * bus/connection.c: maintain separate lists of active and inactive
3840 connections, and a count of each. Maintain count of completed
3841 connections per user. Implement code to check connection limits.
3843 * dbus/dbus-list.c (_dbus_list_unlink): export
3845 * bus/bus.c (bus_context_check_security_policy): enforce a maximum
3846 number of bytes in the message queue for a connection
3848 2003-04-18 Havoc Pennington <hp@pobox.com>
3850 * dbus/dbus-auth.c (record_mechanisms): memleak fixes
3852 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): fix some
3855 * dbus/dbus-keyring.c (add_new_key): fix a memleak, and
3856 on realloc be sure to update the pointer in the keyring
3858 * dbus/dbus-string.c (_dbus_string_zero): compensate for align
3859 offset to avoid writing to unallocated memory
3861 * dbus/dbus-auth.c (process_rejected): return FALSE if we fail to
3862 try the next mechanism, so we properly handle OOM
3864 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): fix double-free
3866 (_dbus_keyring_new): fix OOM bug
3867 (_dbus_keyring_new_homedir): always set error; impose a maximum
3868 number of keys we'll load from the file, mostly to speed up the
3869 test suite and make its OOM checks more useful, but also for
3872 * dbus/dbus-auth.c (process_error_server): reject authentication
3873 if we get an error from the client
3874 (process_cancel): on cancel, send REJECTED, per the spec
3875 (process_error_client): send CANCEL if we get an error from the
3878 2003-04-18 Havoc Pennington <hp@pobox.com>
3880 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix UMR in verbose
3883 * dbus/dbus-auth.c (handle_client_data_cookie_sha1_mech): fix OOM
3886 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): only whine
3887 about DBUS_TEST_HOMEDIR once
3889 * bus/Makefile.am (TESTS_ENVIRONMENT): put DBUS_TEST_HOMEDIR in
3892 * bus/dispatch.c (bus_dispatch_sha1_test): actually load sha1
3893 config file so we test the right thing
3895 Throughout: assorted docs improvements
3897 2003-04-18 Havoc Pennington <hp@pobox.com>
3899 * glib/dbus-gmain.c: adapt to watch changes
3901 * bus/bus.c, bus/activation.c, etc.: adjust to watch changes
3903 * dbus/dbus-server.h: remove dbus_server_handle_watch
3905 * dbus/dbus-connection.h: remove dbus_connection_handle_watch
3907 * dbus/dbus-watch.c (dbus_watch_handle): change DBusWatch to work
3908 like DBusTimeout, so we don't need dbus_connection_handle_watch
3911 2003-04-17 Havoc Pennington <hp@redhat.com>
3913 * dbus/dbus-userdb.c, dbus/dbus-sysdeps.c: redo all the passwd
3914 database usage so it all goes via the DBusUserDatabase cache.
3916 2003-04-17 Havoc Pennington <hp@redhat.com>
3918 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix logic so that if
3919 there was an OOM watch we skipped, we always return TRUE so we
3920 iterate again to have a look at it again. Fixes test suite hang.
3921 Code rearrangement also lets us lose some memset and only iterate
3922 over callbacks once.
3924 * bus/driver.c (bus_driver_handle_message): sense of test for
3927 2003-04-16 Havoc Pennington <hp@pobox.com>
3929 * doc/dbus-specification.sgml: make spec say serials are unsigned
3931 * dbus/dbus-message.h: change message serials to unsigned
3933 * dbus/dbus-connection.c: adapt to message serials being unsigned
3935 2003-04-15 Havoc Pennington <hp@pobox.com>
3937 * bus/bus.c: create and keep around a shared DBusUserDatabase
3940 * bus/connection.c (bus_connection_get_groups): don't cache
3941 groups for user in the connection object, since user database
3942 object now does that.
3944 2003-04-16 Havoc Pennington <hp@redhat.com>
3946 * dbus/dbus-message.c (_dbus_message_add_size_counter): keep a
3947 list of size counters
3948 (_dbus_message_loader_putback_message_link): put back a popped link
3950 * dbus/dbus-connection.c
3951 (dbus_connection_set_max_live_messages_size): rename
3953 (dbus_connection_get_outgoing_size): get size of outgoing
3955 (_dbus_connection_set_connection_counter): remove this cruft
3957 2003-04-14 Havoc Pennington <hp@redhat.com>
3959 * dbus/dbus-userdb.c: user database abstraction, mostly to get
3960 caching, but at some point we might want to be able to use a
3963 * bus/dispatch.c (bus_dispatch_sha1_test): add a test that uses
3964 SHA1 conf file to test the sha1 auth mechanism, since the regular
3965 test always uses EXTERNAL when available.
3968 test/data/valid-config-files/debug-allow-all-sha1.conf.in:
3969 add conf file that requires use of sha1 auth
3971 2003-04-13 Havoc Pennington <hp@pobox.com>
3973 * tools/dbus-send.c, tools/dbus-monitor.c: two utility programs
3974 from Philip Blundell to send messages and monitor them.
3976 2003-04-13 Havoc Pennington <hp@pobox.com>
3978 * dbus/dbus-mainloop.c: fix some reentrancy issues by refcounting
3981 * test/data/valid-config-files/debug-allow-all.conf.in: allow all
3984 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
3985 fix to only recover unused bytes if we're already authenticated
3986 (_dbus_transport_get_is_authenticated): fix to still mark us
3987 authenticated if there are unused bytes.
3989 * bus/dispatch.c: implement security policy checking
3991 * bus/connection.c (bus_transaction_send_from_driver): new
3993 * bus/bus.c (bus_context_check_security_policy): new
3995 * bus/dispatch.c (send_service_nonexistent_error): delete this,
3996 now we just set the DBusError and it gets converted to an error
3999 * bus/connection.c (allow_user_function): enable code using actual
4000 data from the config file
4002 * bus/policy.c (list_allows_user): handle wildcard rules for
4003 user/group connection perms
4005 2003-04-13 Havoc Pennington <hp@pobox.com>
4007 * bus/config-parser.c: Load up the BusPolicy and BusPolicyRules
4009 * dbus/dbus-sysdeps.c (_dbus_get_user_id): new function
4011 * bus/policy.c (bus_policy_append_mandatory_rule)
4012 (bus_policy_append_default_rule, bus_policy_append_user_rule)
4013 (bus_policy_append_group_rule): new functions
4015 2003-04-12 Havoc Pennington <hp@pobox.com>
4017 * bus/config-parser.c (bus_config_parser_new): fix a memleak
4019 * dbus/dbus-sysdeps.c: change DBusCredentials to use longs for
4020 the pid/gid/uid, just for paranoia.
4022 * test/break-loader.c (randomly_do_n_things): find a byte
4023 containing a type code, and randomly change it to a different
4026 2003-04-12 Havoc Pennington <hp@pobox.com>
4028 * bus/policy.h: change BusPolicy to be the thing from the config
4029 file, and rename old BusPolicy to BusClientPolicy
4031 * bus/bus.c, bus/connection.c, bus/config-parser.c: change to
4032 match change in how policy works
4034 * dbus/dbus-internals.h: mark assert_not_reached as
4035 __attribute((noreturn))__
4037 2003-04-11 Havoc Pennington <hp@redhat.com>
4039 * doc/dbus-specification.sgml: fix a spot with the wrong name for
4040 the broadcast service. Use boolean return for ServiceExists.
4042 2003-04-11 Havoc Pennington <hp@redhat.com>
4044 * configure.in: add another directory to look for qt in.
4046 2003-04-11 Havoc Pennington <hp@redhat.com>
4048 * AUTHORS: add Colin Walters
4050 2003-04-11 Havoc Pennington <hp@redhat.com>
4056 2003-04-11 Havoc Pennington <hp@redhat.com>
4058 * bus/messagebus.in: remove pid file when stopping the
4059 message bus, since the bus won't have privileges to remove it
4062 2003-04-11 Havoc Pennington <hp@redhat.com>
4064 * bus/bus.c (bus_context_new): move credentials change after
4067 2003-04-11 Havoc Pennington <hp@pobox.com>
4069 * test/decode-gcov.c: add "below average functions" to the
4070 coverage report, and change how some of the code works.
4072 * bus/test-main.c: bracket some stuff in DBUS_BUILD_TESTS so it's
4073 not in the coverage stats.
4075 * test/test-service.c (main): use _dbus_verbose not fprintf in a
4076 couple places so running the test suite doesn't result in megaspam.
4078 2003-04-11 Havoc Pennington <hp@pobox.com>
4080 * bus/dispatch.c (check_existent_service_activation): accept a no
4081 memory error in a place we didn't before
4083 * bus/test.c (bus_test_run_everything): remove hacky "do it twice
4084 in case the first one failed," since the test suite is less
4087 2003-04-10 Havoc Pennington <hp@pobox.com>
4089 * bus/dispatch.c (check_segfault_service_activation): add test
4090 for launching an executable that just crashes.
4092 * test/test-segfault.c (main): try setting coredumpsize to 0 so we
4093 don't leave a million cores. We'll see how portable this is.
4095 2003-04-10 Havoc Pennington <hp@pobox.com>
4097 * dbus/dbus-spawn.c (_dbus_spawn_async_with_babysitter): move all
4098 the possible parent failures before we fork, so that we don't
4099 fail to create a babysitter after creating the child.
4101 * bus/activation.c (bus_activation_activate_service): kill child
4102 if we don't successfully complete the activation.
4104 2003-04-10 Havoc Pennington <hp@redhat.com>
4106 * dbus/dbus-connection.c (dbus_connection_flush): don't spin on
4107 the connection if it's disconnected
4109 * bus/activation.c (bus_activation_service_created): use new
4110 transaction features to roll back removal of pending activation if
4111 we don't successfully create the service after all. Don't remove
4112 pending activation if the function fails.
4114 * dbus/dbus-list.c (_dbus_list_insert_before_link)
4115 (_dbus_list_insert_after_link): new code to facilitate
4118 * dbus/dbus-hash.c (_dbus_hash_table_insert_string_preallocated):
4119 new functionality, so we can preallocate the ability to insert
4122 * bus/connection.c (bus_transaction_add_cancel_hook): new function
4123 allowing us to put custom hooks in a transaction to be used for
4124 cancelling said transaction
4126 * doc/dbus-specification.sgml: add some discussion of secondary
4127 service owners, and disallow zero-length service names
4129 * bus/services.c (bus_registry_acquire_service): new function,
4130 splits out part of bus_driver_handle_acquire_service() and fixes
4131 a bug where we didn't remove the service doing the acquiring
4132 from the secondary queue if we failed to remove the current owner
4133 from the front of the queue.
4135 2003-04-10 Alexander Larsson <alexl@redhat.com>
4137 * doc/dbus-specification.sgml:
4138 s/org.freedesktop.Broadcast/org.freedesktop.DBus.Broadcast/
4140 2003-04-10 Alexander Larsson <alexl@redhat.com>
4145 Added files to cvsignore
4147 * dbus/dbus-message.h:
4148 * dbus/dbus-message.c: (dbus_message_iter_get_named):
4149 Make get_named() take two out argument and return a boolean.
4150 (dbus_message_iter_get_args_valist):
4151 Update usage of get_named().
4152 (dbus_message_iter_append_byte):
4154 (dbus_message_iter_append_named)
4156 (message_iter_test), (check_message_handling_type), (_dbus_message_test):
4159 2003-04-10 Alexander Larsson <alexl@redhat.com>
4161 * dbus/dbus-marshal.[ch]:
4162 Add array_type_pos argument to _dbus_marshal_validate_arg.
4163 Let you pass a NULL end_pos to _dbus_marshal_validate_type.
4165 * dbus/dbus-message.[ch]:
4166 Multi-dimensional arrays have full type specification in the
4167 outermost array. Iter code re-arranged to handle this.
4168 Added some more iter tests.
4170 * doc/dbus-specification.sgml:
4173 Update new array encoding description.
4174 Correct DBUS_SERVICE_FLAGS_REPLACE_EXISTING description.
4176 * test/data/invalid-messages/array-with-mixed-types.message:
4177 * test/data/valid-messages/array-of-array-of-uint32.message:
4178 Change to the new array format.
4180 * test/data/invalid-messages/too-short-dict.message:
4183 * test/data/valid-messages/recursive-types.message:
4184 Fix up and extend test.
4186 2003-04-10 Havoc Pennington <hp@pobox.com>
4188 * bus/dispatch.c: lots of fixes
4190 * dbus/dbus-mainloop.c (_dbus_loop_dispatch): export
4191 (_dbus_loop_iterate): remove old "quit if no callbacks" code,
4192 that was crack, broke the test service.
4194 * dbus/dbus-transport.c (_dbus_transport_open): fix error
4195 handling to avoid piling up errors if we get a failure on the
4198 * dbus/dbus-internals.c (_dbus_real_assert_not_reached): include
4199 pid in assertion failures.
4201 * dbus/dbus-mainloop.c (_dbus_loop_iterate): use static arrays up
4202 to some fixed size of file descriptor array. Don't return TRUE
4203 anytime a timeout exists, that led to lots of busy loop silliness
4206 2003-04-09 Havoc Pennington <hp@redhat.com>
4208 * dbus/dbus-mainloop.c (check_timeout): fix timeouts, I thought
4209 I'd checked this in earlier but hadn't.
4211 2003-04-09 Havoc Pennington <hp@redhat.com>
4213 * bus/dispatch.c (bus_dispatch_test): get a bit further through
4214 the activation test (man this is getting old!)
4216 2003-04-09 Havoc Pennington <hp@redhat.com>
4218 * test/test-utils.c: use dispatch status function to fix this up
4220 * bus/connection.c (connection_watch_callback): don't dispatch
4222 (connection_timeout_callback): don't dispatch from here
4223 (bus_connections_setup_connection): set the dispatch status function
4224 (bus_connection_disconnected): unset it
4226 * dbus/dbus-mainloop.c (_dbus_loop_queue_dispatch): new function
4227 used to add a connection to be dispatched
4228 (_dbus_loop_iterate): do the dispatching at the end of each
4231 * dbus/dbus-connection.c
4232 (dbus_connection_set_dispatch_status_function): new function
4233 allowing us to fix up main loop usage
4234 (_dbus_connection_last_unref): free all the various function
4236 (dbus_connection_dispatch): call the DispatchStatusFunction
4237 whenever this function returns
4238 (dbus_connection_handle_watch): call DispatchStatusFunction
4239 (dbus_connection_send_with_reply_and_block): call DispatchStatusFunction
4240 (reply_handler_timeout): call DispatchStatusFunction
4241 (dbus_connection_flush): call DispatchStatusFunction
4243 2003-04-09 Havoc Pennington <hp@redhat.com>
4245 * dbus/dbus-bus.c (dbus_bus_register): fix up error handling and
4248 * bus/dispatch.c (check_service_activated): fix bug in test
4250 * dbus/dbus-mainloop.c (check_timeout): fix this up
4252 * dbus/dbus-internals.c (_dbus_verbose_real): include PID in
4253 verbose output so we can sort out output from different processes,
4254 e.g. in the activation case.
4256 2003-04-08 Colin Walters <walters@gnu.org>
4258 * bus/bus.c (struct BusContext) [pidfile]: New member, to store
4260 (bus_context_new): Set it.
4261 (bus_context_unref): Use it to delete the pid file.
4263 2003-04-08 Havoc Pennington <hp@redhat.com>
4265 * test/data/invalid-messages/array-with-mixed-types.message:
4266 regression test that fails for the moment
4268 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): reorder
4269 tests for convenience
4271 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): don't allow
4272 array of nil, it broke things.
4274 * test/data/invalid-messages/array-of-nil.message: regression test
4276 * test/data/valid-messages/array-of-array-of-uint32.message:
4277 happened to write this so added it to suite
4279 2003-04-08 Havoc Pennington <hp@redhat.com>
4281 * bus/driver.c (bus_driver_handle_acquire_service): init
4282 retval/reply before checking name
4284 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add a
4285 recursion depth argument
4287 * dbus/dbus-message.h (struct DBusMessageIter): put some padding
4288 in the public struct for future extension
4290 * dbus/dbus-message-builder.c (_dbus_message_data_load): fix
4293 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix a verbose
4296 * doc/dbus-specification.sgml: fix typo
4298 2003-04-08 Alexander Larsson <alexl@redhat.com>
4300 Implemented recursive types, named types and new-style iters
4303 * glib/test-thread-client.c: (thread_func):
4304 * glib/test-thread-server.c: (handle_test_message):
4305 * test/test-service.c: (handle_echo):
4314 * dbus/dbus-internals.c: (_dbus_type_to_string):
4315 Update for new types.
4317 * dbus/dbus-marshal.[ch]:
4318 Implement recursive types and the new marshalling format.
4319 Remove hardcoded dict marshalling.
4320 Marshal named types.
4322 * dbus/dbus-message-builder.c:
4324 Remove references to old types
4326 * dbus/dbus-message.[ch]:
4327 New non-refcounted iter API that supports recursive iters.
4328 Use iters for appending, including support for recursive
4330 Add byte and named type support.
4331 Update everything to new marshalling formats.
4332 Add tests for new API.
4334 * dbus/dbus-protocol.h:
4335 Remove old array types.
4336 Add types: BYTE, ARRAY, DICT, NAMED
4338 * dbus/dbus-string.c:
4339 * dbus/dbus-sysdeps.c:
4340 Make parse_double locale safe.
4342 * dbus/dbus-test-main.c:
4348 * doc/dbus-specification.sgml:
4351 * test/data/incomplete-messages/missing-body.message:
4352 * test/data/invalid-messages/bad-boolean.message:
4353 * test/data/invalid-messages/bad-boolean-array.message:
4354 * test/data/invalid-messages/boolean-array-length-too-long.message-raw:
4355 * test/data/invalid-messages/boolean-has-no-value.message-raw:
4356 * test/data/invalid-messages/too-short-dict.message:
4357 * test/data/valid-messages/dict-simple.message:
4358 * test/data/valid-messages/dict.message:
4359 * test/data/valid-messages/emptiness.message:
4360 * test/data/valid-messages/lots-of-arguments.message:
4361 * test/data/valid-messages/no-padding.message:
4362 * test/data/valid-messages/recursive-types.message:
4363 Add missing NAME fields
4364 Fix up dicts & arrays
4366 * test/data/invalid-messages/dict-with-nil-value.message:
4367 Removed, this is not invalid anymore.
4369 * test/data/valid-messages/recursive-types.message:
4370 Add new test for deeply recursive types.
4372 2003-04-07 Havoc Pennington <hp@pobox.com>
4374 * bus/driver.c (bus_driver_handle_acquire_service): return an
4375 error if you try to acquire a service that starts with ':'
4377 2003-04-07 Havoc Pennington <hp@redhat.com>
4379 * doc/dbus-specification.sgml: require that base service names
4380 start with ':' and that the base service is created/deleted
4381 as first and last things a connection does on the bus
4383 * bus/dispatch.c (check_existent_service_activation): lots more
4384 work on the activation test; it doesn't fully pass yet...
4386 * test/test-service.c (main): fix so we don't memleak the
4387 connection to the message bus
4388 (filter_func): accept a message asking us to exit
4390 2003-04-06 Havoc Pennington <hp@pobox.com>
4392 * qt/Makefile.am (dbusinclude_HEADERS): install dbus-qt.h,
4395 * configure.in: fixes to Qt detection from Colin Walters
4397 * doc/Makefile.am: Only remove generated docbook dirs if they
4398 exist, from Colin Walters
4400 * dbus/dbus-bus.c: change how we set well-known connections to
4401 NULL, so that it works if a single connection is stored in
4402 two well-known array slots.
4404 * test/Makefile.am: remove a lot of stuff that isn't immediately
4405 useful, it's in CVS history if we want it.
4407 * test/test-service.c: use dbus-mainloop instead of that
4410 2003-04-06 Havoc Pennington <hp@pobox.com>
4412 * dbus/Makefile.am: split lists of sources into stuff that goes in
4413 the library, util functions that go in the lib and are also used
4414 elsewhere, and util functions that are used in tests/daemon but
4415 don't go in the lib.
4417 * dbus/dbus-mainloop.h, dbus/dbus-mainloop.c: move bus/loop.[hc]
4418 here so it can be used in test binaries also
4420 2003-04-06 Havoc Pennington <hp@pobox.com>
4422 * dbus/dbus-sysdeps.c (_dbus_become_daemon): write the pidfile
4423 here in the parent process, so we can return an error if it
4424 fails. Also, move some of the code into the child so the parent
4425 is less hosed if we fail midway through.
4427 * bus/bus.c (bus_context_new): move pidfile detection further up
4428 in the function, before we start overwriting sockets and such.
4430 * bus/messagebus.in: adjust this a bit, not sure if it will work.
4432 * configure.in: add --with-system-pid-file and --with-system-socket
4434 2003-04-06 Colin Walters <walters@verbum.org>
4436 * configure.in (DBUS_SYSTEM_PID_FILE): New variable.
4438 * bus/system.conf.in: Declare a pidfile.
4440 * bus/bus.c (bus_context_new): Test for an existing pid file, and
4441 create one (if appropriate).
4443 * bus/config-parser.c (enum ElementType) [ELEMENT_PIDFILE]: New.
4444 (struct BusConfigParser) [pidfile]: New.
4445 (element_type_to_name, merge_included, start_busconfig_child)
4446 (bus_config_parser_end_element, bus_config_parser_content): Handle it.
4447 (bus_config_parser_unref): Free it.
4448 (bus_config_parser_get_pidfile): New function.
4450 * bus/config-parser.h (_dbus_write_pid_file): Prototype.
4452 * dbus/dbus-errors.h (DBUS_ERROR_PIDFILE_EXISTS): New error.
4454 * dbus/dbus-sysdeps.c (_dbus_write_pid_file): New function.
4456 * dbus/dbus-sysdeps.h: Prototype it.
4458 2003-04-06 Havoc Pennington <hp@pobox.com>
4460 * bus/bus.c (bus_context_new): print the address in here, rather
4461 than in main(), because we need to do it before forking the daemon
4463 * bus/dispatch.c (send_service_nonexistent_error): set the sender
4464 on the service nonexistent error
4466 * bus/driver.c (bus_driver_handle_acquire_service): set the
4467 sender on the AcquireService reply
4469 * test/data/valid-config-files/debug-allow-all.conf.in: Make test
4470 server also listen on a UNIX socket so services can connect to it.
4472 2003-04-06 Havoc Pennington <hp@pobox.com>
4474 * dbus/dbus-threads.c: Redo how the fake debug mutexes are done
4475 so it detects deadlocks and also we actually init threads when
4478 2003-04-06 Havoc Pennington <hp@pobox.com>
4480 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
4481 save the domain socket name, and unlink it when we disconnect the
4482 server. Means that at least when we exit normally, we won't leave
4483 a bunch of junk in /tmp
4485 * dbus/dbus-transport-unix.c
4486 (_dbus_transport_new_for_domain_socket): code cleanup (nicer
4487 memory management). (I was making a real change here but then
4490 2003-04-06 Havoc Pennington <hp@pobox.com>
4492 * bus/bus.c (bus_context_new): fix wrong handling of
4493 server_data_slot_unref() in the error case.
4495 * dbus/dbus-internals.h (_dbus_assert): change so it passes
4496 "(condition) != 0" to _dbus_real_assert so that
4497 "_dbus_assert (pointer)" doesn't cause a warning
4499 * bus/main.c (main): accept --print-address option to print out
4500 the message bus address
4502 * dbus/dbus-sysdeps.c (_dbus_generate_random_ascii): export this
4504 * dbus/dbus-transport.c (_dbus_transport_open): special error for
4505 "tmpdir" option to unix: address on client side
4507 * dbus/dbus-server.c (dbus_server_listen): handle "tmpdir" option
4510 * configure.in (TEST_SOCKET_DIR): locate a temporary directory
4511 we can use to create sockets in the test suite.
4513 * bus/main.c (signal_handler): on SIGTERM, exit the daemon
4514 cleanly. To be used for testing.
4516 * dbus/dbus-spawn.c (babysit): use _dbus_set_signal_handler()
4518 * dbus/dbus-sysdeps.c (_dbus_set_signal_handler): new
4520 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
4521 handle trying to call this when there's no servers active
4523 2003-04-05 Havoc Pennington <hp@pobox.com>
4529 2003-04-05 Havoc Pennington <hp@pobox.com>
4531 * bus/bus.c (setup_server): fix this so dbus-daemon-1 doesn't
4532 crash on startup. Need to get "try starting the daemon"
4533 in the test suite I guess. ;-)
4535 * dbus/dbus-server.h, dbus/dbus-server.c: remove the stuff that
4536 tracked the number of open connections; it's better done in
4537 application-specific code as you want it to span all servers etc.
4539 2003-04-05 Havoc Pennington <hp@pobox.com>
4541 * bus/Makefile.am (install-data-hook): add missing DESTDIR,
4542 patch from Colin Walters
4544 2003-04-05 Havoc Pennington <hp@pobox.com>
4546 * doc/config-file.txt (Elements): fix docs of <auth> to reflect
4547 reality; in fact multiple mechanisms are allowed.
4549 * dbus/dbus-internals.c (_dbus_real_assert)
4550 (_dbus_real_assert_not_reached): move guts of _dbus_assert() and
4551 _dbus_assert_not_reached() into functions, so that they don't show
4552 up in basic block counts for test coverage, and don't use up as
4553 much disk space. Does mean slower execution speed though, so
4554 assumes --disable-asserts is the normal production case.
4556 2003-04-05 Havoc Pennington <hp@pobox.com>
4558 * test/Makefile.am (dist-hook): also dist *.in files
4564 2003-04-05 Havoc Pennington <hp@pobox.com>
4566 * dbus/dbus-string.c: docs warning
4568 * dbus/dbus-spawn.c: missing docs
4570 * dbus/dbus-memory.c (struct ShutdownClosure): missing docs
4572 2003-04-05 Havoc Pennington <hp@pobox.com>
4574 * bus/loop.c (bus_loop_iterate): fix the timeout code, using
4577 * dbus/dbus-spawn.c (_dbus_babysitter_unref): set sitter_pid
4578 to -1 once we've reaped the babysitter
4579 (_dbus_babysitter_handle_watch): do as much work as we can, not
4582 * bus/activation.c: add code using DBusBabysitter so that we
4583 handle it when a service fails to start up properly.
4584 (bus_activation_service_created): don't remove the activation
4585 entries as we go, just let them get removed when we free the pending
4586 activation. Unref reply messages after sending them.
4588 2003-04-05 Havoc Pennington <hp@pobox.com>
4590 * test/decode-gcov.c (main): print per-directory stats in the report
4592 * Makefile.am (coverage-report.txt): don't include test/* in gcov stats
4594 2003-04-05 Havoc Pennington <hp@pobox.com>
4596 * Makefile.am (coverage-report.txt): add target "coverage-report.txt"
4598 * test/decode-gcov.c: hack up a little program to suck data
4599 out of gcov files. Yes this is sort of silly.
4601 * configure.in: define something in config.h and do an
4602 AM_CONDITIONAL when gcov is enabled
4604 2003-04-04 Havoc Pennington <hp@redhat.com>
4606 * dbus/dbus-spawn.c, dbus/dbus-spawn.h: Change dbus_spawn to
4607 return a "babysitter" object that is used to monitor the status of
4608 the spawned process and reap it when required.
4610 * test/test-segfault.c, test/test-exit.c,
4611 test/test-sleep-forever.c: binaries that do various lame things,
4612 used in the test suite.
4614 * dbus/dbus-sysdeps.c: kill _dbus_errno_to_string()
4616 2003-04-03 Havoc Pennington <hp@pobox.com>
4618 * dbus/dbus-spawn.c: Move dbus-spawn into a separate file
4619 in preparation for modifying it, dbus-sysdeps is getting
4622 2003-04-03 Havoc Pennington <hp@redhat.com>
4624 * bus/loop.h, bus/loop.c: make the mainloop an object so we can
4627 * bus/*.[hc]: adapt to mainloop change
4629 2003-04-03 Havoc Pennington <hp@redhat.com>
4631 * bus/activation.c (load_directory): fix up memleaks
4632 (bus_activation_entry_free): free the entry
4634 * dbus/dbus-bus.c (dbus_bus_acquire_service): return an error if
4635 we get one from the message bus; fix memleaks.
4637 * dbus/dbus-message.c (dbus_set_error_from_message): new function
4639 2003-04-03 Havoc Pennington <hp@pobox.com>
4641 * bus/config-parser.c (bus_config_parser_unref): free
4642 list of mechanisms, bug discovered by test suite enhancements
4643 (putting system.conf and session.conf into suite)
4645 * test/Makefile.am, test/test-service.c: add placeholder for a
4646 test service that we'll activate as part of test suite. Doesn't
4649 * dbus/dbus-sysdeps.c (_dbus_setenv): support unsetenv by
4650 setting NULL value, and use system malloc not dbus_malloc()
4651 when we have unavoidable memleakage.
4653 * dbus/dbus-bus.c (dbus_bus_get): fix bug where bus type of 0
4654 didn't work, and support DBUS_BUS_ACTIVATION.
4656 * bus/activation.c (child_setup): pass our well-known bus type to
4659 * bus/config-parser.c: support <type> to specify well-known type
4661 * doc/dbus-specification.sgml: document the env variables to
4662 locate well-known buses and find service activator
4664 2003-04-02 Havoc Pennington <hp@redhat.com>
4666 * test/Makefile.am (all-local): add a rule to copy tests to
4667 builddir, so we can have generated tests. Use this to remove the
4668 silly hack for testing system.conf and session.conf. Will use this
4669 shortly to generate .service files pointing to test binaries.
4671 2003-04-02 Havoc Pennington <hp@redhat.com>
4673 * dbus/dbus-string.c (set_length): fix a bug - we allocated max of
4674 current alloc and needed new length, not max of the doubled
4675 allocation and needed new length. Also, when building tests,
4676 don't do the double-allocation stuff, just realloc every time.
4678 2003-04-02 Havoc Pennington <hp@redhat.com>
4680 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): include filenames
4682 (_dbus_string_get_dirname): new
4683 (_dbus_sysdeps_test): new
4684 (_dbus_directory_open): include dirnames in error messages
4686 * bus/config-parser.c: interpret <include> and <includedir> and
4687 <servicedir> relative to config file location if the given
4688 filename is not absolute.
4690 * dbus/dbus-string.c (_dbus_string_find_byte_backward): new
4692 2003-04-02 Havoc Pennington <hp@redhat.com>
4694 * bus/connection.c (bus_transaction_send_error_reply): set sender
4695 service for the error, and unref the reply on success
4697 * bus/activation.c: convert to use BusTransaction so OOM can be
4699 (bus_activation_service_created): set sender of the message
4701 2003-04-01 Havoc Pennington <hp@redhat.com>
4703 * bus/config-parser.c, bus/bus.c: implement <servicedir> and
4704 <includedir> (at least mostly)
4706 * dbus/dbus-sysdeps.c (_dbus_change_identity): set the group ID
4707 first, then the user ID
4709 2003-04-01 Havoc Pennington <hp@pobox.com>
4711 * dbus/dbus-server.c (dbus_server_set_auth_mechanisms): new
4714 * dbus/dbus-auth.c (_dbus_auth_set_mechanisms): new
4716 * dbus/dbus-internals.c (_dbus_dup_string_array): new function
4718 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): chmod the
4719 socket 0777, and unlink any existing socket.
4721 * bus/bus.c (bus_context_new): change our UID/GID and fork if
4722 the configuration file so specifies; set up auth mechanism
4725 * bus/config-parser.c (bus_config_parser_content): add support
4726 for <fork> option and fill in code for <auth>
4728 * bus/system.conf.in: add <fork/> to default configuration,
4729 and limit auth mechanisms to EXTERNAL
4731 * doc/config-file.txt (Elements): add <fork>
4733 * dbus/dbus-sysdeps.c (_dbus_become_daemon): new function
4734 (_dbus_change_identity): new function
4736 2003-03-31 Havoc Pennington <hp@redhat.com>
4738 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket)
4739 (_dbus_listen_unix_socket): fix off-by-one error in null
4740 termination spotted by Nalin
4742 2003-03-31 Havoc Pennington <hp@redhat.com>
4744 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): allow setting
4745 DBUS_TEST_HOMEDIR when tests are enabled, so we can test without
4746 having a real home directory available.
4748 2003-03-31 Havoc Pennington <hp@redhat.com>
4750 * bus/Makefile.am (install-data-hook): create /var/run/dbus
4752 * bus/messagebus.in: add init script for Red Hat /etc/init.d
4754 * configure.in: add support for specifying a style of init script
4757 2003-03-31 Havoc Pennington <hp@redhat.com>
4759 Fix some annoying DBusString API and fix all affected code.
4761 * dbus/dbus-string.c (_dbus_string_init): get rid of annoying
4763 (_dbus_string_get_data): change to return string instead of using
4765 (_dbus_string_get_const_data): ditto
4766 (_dbus_string_get_data_len): ditto
4767 (_dbus_string_get_const_data_len): ditto
4769 2003-03-31 Havoc Pennington <hp@redhat.com>
4771 * bus/main.c (main): fix up the command line arguments to be nicer
4773 2003-03-31 Havoc Pennington <hp@redhat.com>
4775 * dbus/Makefile.am (INCLUDES): use EXPANDED_LOCALSTATEDIR to
4776 define DBUS_SYSTEM_BUS_PATH as we want to compile in the same
4777 final location that lands in the config file
4779 * bus/config-loader-expat.c (bus_config_load): fix type of
4782 * doc/TODO: remove TODO item for dbus_bus_get()
4784 * dbus/dbus-bus.c (bus_data_free): add missing lock/unlock
4786 2003-03-31 Havoc Pennington <hp@pobox.com>
4788 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_domain_socket)
4789 (_dbus_transport_new_for_tcp_socket): these didn't need the "server"
4790 argument since they are always client side
4792 * dbus/dbus-server.c (dbus_server_get_address): new function
4794 * bus/main.c (main): take the configuration file as an argument.
4796 * test/data/valid-config-files/debug-allow-all.conf: new file to
4797 use with dispatch.c tests for example
4799 * bus/test-main.c (main): require test data dir
4801 * bus/bus.c (bus_context_new): change this to take a
4802 configuration file name as argument
4804 * doc/config-file.txt (Elements): add <servicedir>
4806 * bus/system.conf, bus/session.conf: new files
4808 * dbus/dbus-bus.c (dbus_bus_get): look for system bus on
4809 well-known socket if none set
4811 * configure.in: create system.conf and session.conf
4813 2003-03-30 Havoc Pennington <hp@pobox.com>
4815 * bus/config-parser.c: hacking
4817 * dbus/dbus-memory.c: don't use DBusList for the list of stuff
4818 to shut down, since it could cause weirdness with the DBusList
4821 * dbus/dbus-list.c (_dbus_list_test): add tests for the
4822 link-oriented stack routines
4823 (alloc_link): free the mempool if the first alloc from it fails
4825 * dbus/dbus-mempool.c (struct DBusMemBlock): fix alignment issue
4827 * dbus/dbus-string.c (UNICODE_VALID): sync new version of this
4829 (_dbus_string_skip_white): new
4831 * doc/config-file.txt (Elements): add <includedir>
4833 2003-03-28 Havoc Pennington <hp@pobox.com>
4835 * dbus/dbus-string.c (_dbus_string_copy_data_len)
4836 (_dbus_string_copy_data): new functions
4838 2003-03-28 Anders Carlsson <andersca@codefactory.se>
4840 * dbus/dbus-bus.c: (bus_data_free), (dbus_bus_get):
4844 * dbus/dbus-memory.c:
4847 2003-03-28 Havoc Pennington <hp@pobox.com>
4849 * bus/test.c (bus_test_flush_bus): remove the sleep from here,
4850 I think it may have just been superstition. Not sure.
4852 * dbus/dbus-string.c (_dbus_string_base64_decode): catch some OOM
4853 failures that were not being handled.
4855 * dbus/dbus-auth.c (process_auth): fix a memleak in OOM handling
4857 * dbus/dbus-memory.c: add ability to set number of mallocs in a
4858 row that will fail on out-of-memory.
4860 * dbus/dbus-internals.c (_dbus_test_oom_handling): convenience
4861 function for testing out-of-memory handling.
4863 * bus/config-loader-expat.c (memsuite): don't wrap the dbus
4864 allocation functions, they do map exactly to the expat ones.
4866 2003-03-27 Havoc Pennington <hp@redhat.com>
4868 * bus/config-loader-libxml.c (bus_config_load): add another error
4871 2003-03-26 Anders Carlsson <andersca@codefactory.se>
4874 Add note about automatic service activation.
4876 * doc/dbus-specification.sgml:
4877 Rename the specification and clarify a few things.
4879 2003-03-26 Anders Carlsson <andersca@codefactory.se>
4882 * dbus/dbus-address.c:
4884 * dbus/dbus-marshal.c:
4885 * dbus/dbus-server-debug-pipe.c:
4886 * dbus/dbus-transport-unix.c:
4887 Fix documentation warnings.
4889 2003-03-26 Havoc Pennington <hp@pobox.com>
4891 * bus/test-main.c, dbus/dbus-test.c (main): check memleaks
4892 after every test so it's quick and easy to see which leaked, and
4893 so we test multiple dbus_shutdown() calls
4895 * configure.in: change configure.in XML stuff to also support
4898 * config-loader-libxml.c: some hacking
4900 * config-loader-expat.c: some hacking
4902 * config-parser.c: some hacking, plus tests
4904 2003-03-25 Havoc Pennington <hp@redhat.com>
4906 * throughout - add more _DBUS_ASSERT_ERROR_IS_CLEAR
4908 * configure.in: add --with-xml option to specify XML library,
4909 right now only libxml is supported.
4911 * bus/config-loader-libxml.c, config-parser.c: sync some minor
4912 nonworking code between home and work, still just stubs
4914 2003-03-24 Havoc Pennington <hp@redhat.com>
4916 * dbus/dbus-sysdeps.c (_dbus_set_fd_nonblocking): move to this
4919 * dbus/dbus-errors.c (dbus_set_error, dbus_set_error_const): allow
4920 NULL argument for "message" if the error is a well-known one,
4921 fill in a generic message in this case.
4923 * dbus/dbus-errors.h (DBusResultCode): Kill DBusResultCode in
4926 * bus/test.c (bus_test_flush_bus): add
4928 * bus/policy.c (bus_policy_test): test code stub
4930 2003-03-24 Havoc Pennington <hp@pobox.com>
4932 * bus/connection.c (bus_connections_setup_connection): set up
4933 the "can this user connect" function, but it always returns
4934 TRUE until we have a config file parser so we can have a config
4935 file that allows connections.
4937 2003-03-23 Havoc Pennington <hp@pobox.com>
4939 * dbus/dbus-threads.c (dbus_mutex_new, dbus_condvar_new): with
4940 DBUS_BUILD_TESTS, actually alloc/free a block of memory for
4941 the mutex, so we can check for proper memory management
4944 * dbus/dbus-dataslot.c: remove the mutex from
4945 DBusDataSlotAllocator and lock it manually when using it,
4946 to simplify fitting it into the global slots framework.
4948 * dbus/dbus-threads.c (init_static_locks): rework how we're
4949 handling global locks so they are easily shut down.
4951 * bus/policy.c (bus_policy_append_rule): fix
4953 * bus/test-main.c (main): check for memleaks
4955 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): make
4956 test suite check for memleaks
4958 * dbus/dbus-memory.c: add support in test mode for tracking
4959 number of outstanding blocks
4961 2003-03-23 Havoc Pennington <hp@pobox.com>
4963 * bus/policy.c, bus/bus.c, bus/connection.c: implement allow/deny
4966 * dbus/dbus-hash.h: add ULONG hash keys
4968 * dbus/dbus-sysdeps.c (_dbus_get_groups): new
4969 (_dbus_get_group_id): new function
4971 2003-03-20 Havoc Pennington <hp@redhat.com>
4973 * dbus/dbus-connection.c (dbus_connection_set_unix_user_function):
4975 (dbus_connection_get_unix_user): new function
4977 2003-03-20 Havoc Pennington <hp@pobox.com>
4979 * bus/connection.c (bus_connection_send_oom_error): assert that
4980 message has a sender
4981 (connection_execute_transaction): ditto
4982 (bus_connection_preallocate_oom_error): fix to set the sender, and
4983 set recipient to the destination service, not the bus driver
4985 * bus/policy.c: hacking
4987 * dbus/dbus-message.c (dbus_message_service_is): new function
4988 (dbus_message_sender_is): new
4990 2003-03-19 Havoc Pennington <hp@redhat.com>
4992 * bus/policy.c: start sketching code for policy restrictions on
4993 what connections can do.
4995 2003-03-18 Havoc Pennington <hp@redhat.com>
4997 * doc/TODO: some notes on high-level todo items. Little nitpick
4998 stuff is all in @todo, so no need to add it here.
5000 * doc/config-file.txt: some notes on how config file might look
5002 2003-03-18 Anders Carlsson <andersca@codefactory.se>
5008 2003-03-17 Havoc Pennington <hp@redhat.com>
5010 * dbus/dbus-internals.h: add gcc attributes so that
5011 our printf-style functions warn on bad arguments to
5014 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): fix printf
5017 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
5020 2003-03-17 Havoc Pennington <hp@redhat.com>
5022 * bus/test-main.c (main): make it print something as it runs
5023 so make check doesn't look stuck
5025 * doc/negotiation.txt, doc/dbus-sasl-profile.txt: remove
5026 from CVS, now obsolete
5028 2003-03-17 Anders Carlsson <andersca@codefactory.se>
5030 * bus/dispatch.c: (bus_dispatch):
5031 Refetch the service name since it may have been reallocated
5032 when dbus_message_set_sender was called.
5034 * dbus/dbus-sysdeps.c: (_dbus_accept):
5035 Add address and address length variables and use them to stop
5036 valgrind from complaining.
5038 2003-03-17 Havoc Pennington <hp@pobox.com>
5040 All tests pass, no memleaks, no valgrind complaints.
5042 * bus/test.c: refcount handler_slot
5044 * bus/connection.c (bus_connections_new): refcount
5045 connection_data_slot
5047 * dbus/dbus-auth-script.c (_dbus_auth_script_run): delete unused
5048 bytes so that auth scripts pass.
5050 * bus/dispatch.c: init message_handler_slot so it gets allocated
5053 * bus/dispatch.c (message_handler_slot_ref): fix memleak
5055 * dbus/dbus-server-debug-pipe.c (_dbus_server_debug_pipe_new):
5056 dealloc server_pipe_hash when no longer used for benefit of
5059 * dbus/dbus-auth.c (process_command): memleak fix
5061 * bus/dispatch.c (check_hello_message): memleak fix
5063 2003-03-16 Havoc Pennington <hp@pobox.com>
5065 * dbus/dbus-bus.c (ensure_bus_data): fix double-unref of the data slot
5067 2003-03-17 Anders Carlsson <andersca@codefactory.se>
5069 * bus/activation.c (bus_activation_activate_service): Append
5070 the pending activation entry to the list of pending activations.
5072 2003-03-16 Havoc Pennington <hp@pobox.com>
5074 * bus/dispatch.c (bus_dispatch_test): remove double-unrefs of
5077 * dbus/dbus-address.c (create_entry): fix OOM handling when
5078 failing to alloc entry->method
5080 2003-03-16 Havoc Pennington <hp@pobox.com>
5082 * dbus/dbus-watch.c (_dbus_watch_new): handle failure to malloc
5085 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
5086 add some missing dbus_set_result
5088 * bus/dispatch.c (bus_dispatch_add_connection): handle failure to
5089 alloc the DBusMessageHandler
5091 * dbus/dbus-transport.c (_dbus_transport_disconnect): don't ref
5092 the transport here, since we call this from the finalizer; it
5093 resulted in a double-finalize.
5095 * dbus/dbus-transport.c (_dbus_transport_disconnect): fix a bug
5096 where we tried to use transport->connection that was NULL,
5097 happened when transport was disconnected early on due to OOM
5099 * bus/*.c: adapt to handle OOM for watches/timeouts
5101 * dbus/dbus-transport-unix.c: port to handle OOM during
5104 * dbus/dbus-auth.c (_dbus_auth_get_unused_bytes): return a
5105 reference to unused bytes instead of a copy
5107 * dbus/dbus-server.c (dbus_server_handle_watch): return FALSE for
5110 * dbus/dbus-connection.c (dbus_connection_handle_watch): return
5113 * dbus/dbus-timeout.c (dbus_timeout_handle): return FALSE for out
5116 2003-03-16 Anders Carlsson <andersca@codefactory.se>
5118 * doc/dbus-specification.sgml:
5119 Document reply message for ActivateService.
5121 2003-03-16 Anders Carlsson <andersca@codefactory.se>
5123 * bus/activation.c: (bus_pending_activation_entry_free),
5124 (bus_pending_activation_free), (bus_activation_new),
5125 (bus_activation_unref), (bus_activation_service_created),
5126 (bus_activation_activate_service):
5128 * bus/bus.c: (bus_context_new):
5129 * bus/desktop-file.c: (new_section):
5130 * bus/driver.c: (bus_driver_send_service_deleted),
5131 (bus_driver_handle_activate_service):
5132 * bus/services.c: (bus_registry_new), (bus_registry_ensure):
5134 * dbus/dbus-connection.c:
5135 (dbus_connection_send_with_reply_and_block):
5136 * dbus/dbus-message.c: (dbus_message_append_args_valist):
5137 * dbus/dbus-protocol.h:
5138 Make activation work better. Now pending activations will be queued
5139 and the daemon won't try to activate services that are already registered.
5141 2003-03-16 Havoc Pennington <hp@pobox.com>
5143 * dbus/dbus-bus.c (ensure_bus_data): handle failure to set
5146 * dbus/dbus-memory.c (_dbus_initialize_malloc_debug): support
5147 DBUS_MALLOC_BACKTRACES to print trace when failing an alloc
5149 2003-03-16 Havoc Pennington <hp@pobox.com>
5151 * dbus/dbus-string.c (_dbus_string_validate_utf8): oops, unbreak
5152 this. always run the test suite before commit...
5154 * bus/*: adapt to DBusConnection API changes
5156 * glib/dbus-gmain.c: adapt to DBusConnection API changes,
5157 requires renaming stuff to avoid dbus_connection_dispatch name
5160 * dbus/dbus-transport.c (_dbus_transport_queue_messages): new
5163 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
5164 separate from _dbus_message_loader_return_buffer()
5166 * dbus/dbus-connection.c (dbus_connection_get_n_messages): remove
5167 this, because it's now always broken to use; the number of
5168 messages in queue vs. the number still buffered by the message
5169 loader is undefined/meaningless. Should use
5170 dbus_connection_get_dispatch_state().
5171 (dbus_connection_dispatch): rename from
5172 dbus_connection_dispatch_message
5174 2003-03-16 Havoc Pennington <hp@pobox.com>
5176 * dbus/dbus-string.c (_dbus_string_validate_utf8): copy in a real
5179 2003-03-16 Anders Carlsson <andersca@codefactory.se>
5181 * dbus/dbus-connection.c:
5182 (dbus_connection_send_with_reply_and_block):
5183 Decrease connection->n_incoming when removing an entry
5185 * dbus/dbus-dict.c: (dbus_dict_entry_free),
5186 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
5187 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
5188 (dbus_dict_set_byte_array), (dbus_dict_set_string_array),
5189 (dbus_dict_get_boolean_array), (dbus_dict_get_double_array),
5190 (dbus_dict_get_byte_array):
5191 Handle NULL arrays and strings. Also add support for byte arrays.
5193 * dbus/dbus-marshal.c: (_dbus_marshal_byte_array),
5194 (_dbus_marshal_dict), (_dbus_demarshal_byte_array),
5195 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
5196 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
5197 (_dbus_demarshal_dict), (demarshal_and_validate_len),
5198 (_dbus_marshal_validate_arg), (_dbus_marshal_test):
5199 * dbus/dbus-marshal.h:
5200 Add support for marshalling and demarshalling empty arrays and strings.
5202 * dbus/dbus-message.c: (dbus_message_append_args_valist),
5203 (dbus_message_append_string_array),
5204 (dbus_message_iter_get_boolean),
5205 (dbus_message_iter_get_boolean_array),
5206 (dbus_message_iter_get_int32_array),
5207 (dbus_message_iter_get_uint32_array),
5208 (dbus_message_iter_get_double_array),
5209 (dbus_message_iter_get_byte_array),
5210 (dbus_message_iter_get_string_array), (dbus_message_iter_get_dict),
5211 (check_message_handling):
5212 Add support for getting empty arrays and dicts.
5214 * dbus/dbus-string.c: (_dbus_string_validate_utf8):
5215 Don't do any validation at all for now, that's better than just checking
5218 * test/data/valid-messages/emptiness.message:
5219 New test message with lots of empty arrays.
5221 2003-03-16 Havoc Pennington <hp@pobox.com>
5223 * dbus/dbus-connection.c
5224 (_dbus_connection_queue_received_message_link): new function that
5225 can't fail due to OOM
5227 * dbus/dbus-message.c (_dbus_message_loader_pop_message_link):
5228 new function pops a message together with a list link
5231 * dbus/dbus-transport-unix.c (queue_messages): use new link-based
5232 message queuing functions to avoid needing to alloc memory
5234 2003-03-16 Havoc Pennington <hp@pobox.com>
5236 Oops - test code was only testing failure of around 30 of the
5237 mallocs in the test path, but it turns out there are 500+
5238 mallocs. I believe this was due to misguided linking setup such
5239 that there was one copy of dbus_malloc etc. in the daemon and one
5240 in the shared lib, and only daemon mallocs were tested. In any
5241 case, the test case now tests all 500+ mallocs, and doesn't pass
5242 yet, though there are lots of fixes in this patch.
5244 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
5245 this so that it doesn't need to allocate memory, since it
5246 has no way of indicating failure due to OOM (and would be
5247 annoying if it did).
5249 * dbus/dbus-list.c (_dbus_list_pop_first_link): new function
5251 * bus/Makefile.am: rearrange to create two self-contained
5252 libraries, to avoid having libraries with overlapping symbols.
5253 that was resulting in weirdness, e.g. I'm pretty sure there
5254 were two copies of global static variables.
5256 * dbus/dbus-internals.c: move the malloc debug stuff to
5259 * dbus/dbus-list.c (free_link): free list mempool if it becomes
5262 * dbus/dbus-memory.c (_dbus_disable_mem_pools): new function
5264 * dbus/dbus-address.c (dbus_parse_address): free list nodes
5267 * bus/dispatch.c (bus_dispatch_add_connection): free
5268 message_handler_slot when no longer using it, so
5269 memory leak checkers are happy for the test suite.
5271 * dbus/dbus-server-debug-pipe.c (debug_finalize): free server name
5273 * bus/bus.c (new_connection_callback): disconnect in here if
5274 bus_connections_setup_connection fails.
5276 * bus/connection.c (bus_connections_unref): fix to free the
5278 (bus_connections_setup_connection): if this fails, don't
5279 disconnect the connection, just be sure there are no side
5282 * dbus/dbus-string.c (undo_alignment): unbreak this
5284 * dbus/dbus-auth.c (_dbus_auth_unref): free some stuff we were
5286 (_dbus_auth_new): fix the order in which we free strings
5289 * bus/connection.c (bus_connection_disconnected): fix to
5290 not send ServiceDeleted multiple times in case of memory
5293 * dbus/dbus-bus.c (dbus_bus_get_base_service): new function to
5294 get the base service name
5295 (dbus_bus_register_client): don't return base service name,
5296 instead store it on the DBusConnection and have an accessor
5298 (dbus_bus_register_client): rename dbus_bus_register()
5300 * bus/dispatch.c (check_hello_message): verify that other
5301 connections on the bus also got the correct results, not
5302 just the one sending hello
5304 2003-03-15 Havoc Pennington <hp@pobox.com>
5306 Make it pass the Hello handling test including all OOM codepaths.
5307 Now to do other messages...
5309 * bus/services.c (bus_service_remove_owner): fix crash when
5310 removing owner from an empty list of owners
5311 (bus_registry_ensure): don't leave service in the list of
5312 a connection's owned services if we fail to put the service
5315 * bus/connection.c (bus_connection_preallocate_oom_error): set
5316 error flag on the OOM error.
5318 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
5319 handle _dbus_transport_set_connection failure
5321 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd): modify
5322 to create watches up front and simply enable/disable them as
5324 (unix_connection_set): this can now fail on OOM
5326 * dbus/dbus-timeout.c, dbus/dbus-watch.c: add concept
5327 of enabling/disabling a watch or timeout.
5329 * bus/loop.c (bus_loop_iterate): don't touch disabled
5332 * glib/dbus-gmain.c: adapt to enable/disable watches and timeouts
5334 2003-03-15 Havoc Pennington <hp@pobox.com>
5336 * bus/dispatch.c (bus_dispatch_test): OK, now finally actually
5337 write useful test code, after all that futzing around ;-)
5339 Test does not yet pass because we can't handle OOM in
5340 _dbus_transport_messages_pending (basically,
5341 dbus_connection_preallocate_send() does not prealloc the write
5342 watch). To fix this, I think we need to add new stuff to
5343 set_watch_functions, namely a SetEnabled function so we can alloc
5344 the watch earlier, then enable it later.
5346 * dbus/Makefile.am (libdbus_convenience_la_SOURCES): move
5347 dbus-memory.c to the convenience lib
5349 * bus/test.c: rename some static functions to keep them clearly
5350 distinct from stuff in connection.c. Handle client disconnection.
5352 2003-03-14 Havoc Pennington <hp@pobox.com>
5354 * bus/dispatch.c (bus_dispatch_test): do test using debug-pipe
5355 transport, tests more of the real codepath. Set up clients
5356 with bus_setup_debug_client.
5358 * bus/test.c (bus_setup_debug_client): function to set up debug
5359 "clients" on the main loop
5361 * dbus/dbus-transport.c (_dbus_transport_open): add debug-pipe
5364 * dbus/dbus-server.c (dbus_server_listen): add debug-pipe
5367 * dbus/dbus-server-debug.c: support a debug server based on pipes
5369 * dbus/dbus-sysdeps.c (_dbus_full_duplex_pipe): new function
5370 (_dbus_close): new function
5372 * configure.in: check for socketpair
5374 2003-03-14 Havoc Pennington <hp@redhat.com>
5376 * dbus/dbus-memory.c: add a "detect buffer overwrites on free"
5379 * dbus/dbus-transport-debug.c: rework this a good bit to be
5380 less complicated. hopefully still works.
5382 * dbus/dbus-server-debug.c (handle_new_client): remove timeout
5385 * glib/dbus-gmain.c (timeout_handler): don't remove timeout
5388 * dbus/dbus-message.c (dbus_message_copy): rename from
5389 dbus_message_new_from_message, fix it up to copy
5390 all the message fields, add test case
5392 * bus/dispatch.c (bus_dispatch_test): add some more test code,
5393 not quite passing yet
5395 2003-03-14 Havoc Pennington <hp@pobox.com>
5397 * bus/loop.c (bus_loop_iterate): add this so we can "run loop
5398 until no work remains" in test code. (the large diff here
5399 is just code movement, no actual changes)
5401 * dbus/dbus-server-debug.c (DEFAULT_INTERVAL): change interval to
5402 1, no point waiting around for test code.
5403 (_dbus_server_debug_accept_transport): unref the timeout
5404 after adding it (right?)
5406 * dbus/dbus-transport-debug.c (DEFAULT_INTERVAL): ditto
5408 2003-03-13 Havoc Pennington <hp@redhat.com>
5410 * dbus/dbus-timeout.c (_dbus_timeout_list_set_functions): handle
5413 * dbus/dbus-watch.c (_dbus_watch_list_set_functions): handle out
5416 * dbus/dbus-connection.h: Make AddWatchFunction and
5417 AddTimeoutFunction return a bool so they can fail on out-of-memory
5419 * bus/bus.c (bus_context_new): set up timeout handlers
5421 * bus/connection.c (bus_connections_setup_connection): set up
5424 * glib/dbus-gmain.c: adapt to the fact that set_functions stuff
5427 * bus/bus.c (bus_context_new): adapt to changes
5429 * bus/connection.c: adapt to changes
5431 * test/watch.c: adapt to DBusWatch changes
5433 * bus/dispatch.c (bus_dispatch_test): started adding this but
5436 2003-03-14 Anders Carlsson <andersca@codefactory.se>
5438 * bus/dispatch.c (send_service_nonexistent_error): Fix typo.
5440 2003-03-13 Havoc Pennington <hp@pobox.com>
5442 * bus/test.c, bus/test.h, bus/Makefile.am, bus/test-main.c:
5443 set up a test framework as for the library
5445 2003-03-12 Havoc Pennington <hp@pobox.com>
5447 Throughout: purge global variables, introduce BusActivation,
5448 BusConnections, BusRegistry, etc. objects instead.
5450 * bus/bus.h, bus/bus.c: introduce BusContext as a global
5453 * test/Makefile.am (TEST_BINARIES): disable bus-test for now,
5454 going to redo this a bit differently I think
5456 2003-03-12 Havoc Pennington <hp@redhat.com>
5458 Mega-patch that gets the message bus daemon initially handling
5459 out-of-memory. Work still needed. Also lots of random
5460 moving stuff to DBusError instead of ResultCode.
5462 * dbus/dbus-list.c (_dbus_list_length_is_one): new function
5464 * dbus/dbus-connection.c
5465 (dbus_connection_send_with_reply_and_block): use DBusError
5467 * dbus/dbus-bus.c: adapt to API changes, make it use DBusError not
5470 * dbus/dbus-connection.c (dbus_connection_send): drop the result
5471 code here, as the only failure possible is OOM.
5473 * bus/connection.c (bus_connection_disconnect):
5474 rename bus_connection_disconnected as it's a notification only
5476 * bus/driver.c (bus_driver_handle_acquire_service): don't free
5477 "name" on get_args failure, should be done by get_args;
5478 don't disconnect client for bad args, just return an error.
5479 (bus_driver_handle_service_exists): ditto
5481 * bus/services.c (bus_services_list): NULL-terminate returned array
5483 * bus/driver.c (bus_driver_send_service_lost)
5484 (bus_driver_send_service_acquired): send messages from driver to a
5485 specific client to the client's unique name, not to the broadcast
5488 * dbus/dbus-message.c (decode_header_data): reject messages that
5489 contain no name field
5490 (_dbus_message_get_client_serial): rename to
5491 dbus_message_get_serial and make public
5492 (_dbus_message_set_serial): rename from set_client_serial
5493 (_dbus_message_set_reply_serial): make public
5494 (_dbus_message_get_reply_serial): make public
5496 * bus/connection.c (bus_connection_foreach): allow stopping
5497 iteration by returning FALSE from foreach function.
5499 * dbus/dbus-connection.c (dbus_connection_send_preallocated)
5500 (dbus_connection_free_preallocated_send)
5501 (dbus_connection_preallocate_send): new API for sending a message
5502 without possibility of malloc failure.
5503 (dbus_connection_send_message): rename to just
5504 dbus_connection_send (and same for whole function family)
5506 * dbus/dbus-errors.c (dbus_error_free): make this reinit the error
5508 * dbus/dbus-sysdeps.c (_dbus_exit): new function
5510 * bus/activation.c: handle/return errors
5512 * dbus/dbus-errors.h: add more DBUS_ERROR #define
5514 * dbus/dbus-sysdeps.c (_dbus_directory_open) (_dbus_file_get_contents)
5515 (_dbus_directory_get_next_file): use DBusError instead of DBusResultCode
5516 (_dbus_result_from_errno): move to this file
5518 2003-03-10 Anders Carlsson <andersca@codefactory.se>
5520 * dbus/dbus-marshal.c:
5521 (_dbus_marshal_set_string):
5522 Take a length argument so we can marshal the correct string
5525 (_dbus_marshal_dict), (_dbus_demarshal_dict),
5526 (_dbus_marshal_get_arg_end_pos), (_dbus_marshal_validate_arg),
5527 (_dbus_marshal_test):
5528 * dbus/dbus-marshal.h:
5529 Add support for marshalling and demarshalling dicts.
5531 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
5532 Add support for TYPE DICT.
5534 * dbus/dbus-message.c: (set_string_field):
5535 Adjust header padding.
5537 (dbus_message_append_args_valist), (dbus_message_append_dict),
5538 (dbus_message_get_args_valist), (dbus_message_iter_get_arg_type),
5539 (dbus_message_iter_get_dict), (_dbus_message_loader_return_buffer),
5540 (check_message_handling), (check_have_valid_message):
5541 * dbus/dbus-message.h:
5542 Add functions for setting and getting dicts.
5544 * dbus/dbus-protocol.h:
5550 * doc/dbus-specification.sgml:
5551 Add information about how dicts are marshalled.
5553 * test/data/invalid-messages/dict-with-nil-value.message:
5554 * test/data/invalid-messages/too-short-dict.message:
5555 * test/data/valid-messages/dict-simple.message:
5556 * test/data/valid-messages/dict.message:
5557 Add sample messages containing dicts.
5559 2003-03-08 Anders Carlsson <andersca@codefactory.se>
5561 * dbus/dbus-dict.h: Add DBUS_END_DECLS.
5563 2003-03-07 Anders Carlsson <andersca@codefactory.se>
5566 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
5567 (dbus_dict_get_keys), (insert_entry), (dbus_dict_set_boolean),
5568 (dbus_dict_set_int32), (dbus_dict_set_uint32),
5569 (dbus_dict_set_double), (dbus_dict_set_string),
5570 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
5571 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
5572 (dbus_dict_set_string_array), (_dbus_dict_test):
5574 Fix according to comments from Havoc.
5576 2003-03-06 Michael Meeks <michael@server.home>
5578 * configure.in: if we don't have kde-config, disable have_qt.
5580 2003-03-07 Anders Carlsson <andersca@codefactory.se>
5585 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
5586 (dbus_dict_ref), (dbus_dict_unref), (dbus_dict_contains),
5587 (dbus_dict_remove), (dbus_dict_get_value_type),
5588 (dbus_dict_get_keys), (dbus_dict_put_boolean),
5589 (dbus_dict_put_int32), (dbus_dict_put_uint32),
5590 (dbus_dict_put_double), (dbus_dict_put_string),
5591 (dbus_dict_put_boolean_array), (dbus_dict_put_int32_array),
5592 (dbus_dict_put_uint32_array), (dbus_dict_put_double_array),
5593 (dbus_dict_put_string_array), (dbus_dict_get_boolean),
5594 (dbus_dict_get_int32), (dbus_dict_get_uint32),
5595 (dbus_dict_get_double), (dbus_dict_get_string),
5596 (dbus_dict_get_boolean_array), (dbus_dict_get_int32_array),
5597 (dbus_dict_get_uint32_array), (dbus_dict_get_double_array),
5598 (dbus_dict_get_string_array), (_dbus_dict_test):
5600 Add DBusDict implementation
5602 * dbus/dbus-test.c: (dbus_internal_do_not_use_run_tests):
5606 2003-03-04 Havoc Pennington <hp@pobox.com>
5608 * test/data/auth/*: adapt to changes
5610 * dbus/dbus-auth-script.c (_dbus_auth_script_run): add
5611 USERID_BASE64 and change USERNAME_BASE64 to put in username not
5614 * dbus/dbus-keyring.c (_dbus_keyring_validate_context): prevent
5615 more stuff from being in a context name, to make the protocol
5616 simpler to deal with
5618 * dbus/dbus-errors.c (dbus_error_has_name): new function
5619 (dbus_error_is_set): new function
5621 * dbus/dbus-auth.c: replace DBUS_STUPID_TEST_MECH auth
5622 with DBUS_COOKIE_SHA1, implement DBUS_COOKIE_SHA1
5624 * dbus/dbus-connection.c (dbus_connection_flush): also read
5625 messages during a flush operation
5627 * dbus/Makefile.am: remove dbus-md5 since it isn't currently used.
5629 2003-03-05 Anders Carlsson <andersca@codefactory.se>
5631 * configure.in: Check for gethostbyname on Solaris.
5633 * dbus/dbus-transport.c: (_dbus_transport_open):
5634 Remove duplicate "tcp" entry.
5636 * doc/dbus-specification.sgml:
5637 Clarify some things.
5639 2003-03-05 Anders Carlsson <andersca@codefactory.se>
5641 * dbus/dbus-auth.c: (send_rejected), (process_test_subdir):
5642 * dbus/dbus-keyring.c: (_dbus_keyring_new_homedir),
5643 (_dbus_keyring_test):
5644 * dbus/dbus-md5.c: (_dbus_md5_compute):
5645 * dbus/dbus-sha.c: (_dbus_sha_compute):
5648 2003-03-05 Anders Carlsson <andersca@codefactory.se>
5650 * README: Add some things.
5652 2003-03-04 Anders Carlsson <andersca@codefactory.se>
5654 * dbus/dbus-message.c (dbus_message_append_args_valist): Add a break;
5655 after case DBUS_TYPE_BOOELAN.
5657 2003-03-02 Havoc Pennington <hp@pobox.com>
5659 * test/break-loader.c (randomly_set_extreme_ints): add test that
5660 sets really huge and small integers
5662 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add check
5663 that length of boolean array fits in the string, and that
5664 string has room for boolean value in single-bool case.
5666 * dbus/dbus-message-builder.c (_dbus_message_data_load): add
5667 optional value to "ALIGN" command which is what to fill the
5670 * test/data/valid-messages/no-padding.message: add regression
5671 test for the message padding problem
5673 2003-03-02 Havoc Pennington <hp@pobox.com>
5675 * dbus/dbus-message.c (decode_header_data): fix to always init
5676 message_padding, from Benjamin Dauvergne
5678 2003-03-02 Havoc Pennington <hp@pobox.com>
5684 2003-03-01 Joe Shaw <joe@assbarn.com>
5686 * configure.in: Check for "struct cmsgcred" and try to access its
5687 members for BSD-like unices.
5689 * dbus/dbus-sysdeps.c (read_credentials_byte): Fold this back into
5690 _dbus_read_credentials_unix_socket().
5691 (_dbus_read_credentials_unix_socket): Use recvmsg() instead of
5692 read() for reading the credential byte off the unix socket. Use
5693 struct cmsgcred on systems that support it.
5695 2003-02-27 Alexander Larsson <alexl@redhat.com>
5699 Make gthreads-2.0 dependency optional. Don't build thread test if
5702 2003-02-27 Havoc Pennington <hp@pobox.com>
5704 * dbus/dbus-connection.c
5705 (dbus_connection_send_message_with_reply_and_block): fix doh!
5706 doh! doh! bug that resulted in never removing a reply from the
5707 queue, no wonder we called get_reply_serial so much ;-)
5709 * dbus/dbus-message.c (struct DBusMessage): cache reply serial
5710 and client serial instead of demarshaling them every time
5712 2003-02-27 Havoc Pennington <hp@pobox.com>
5714 * dbus/dbus-marshal.c (_dbus_demarshal_int32): rewrite to be much
5715 more inlined, using dbus-string-private.h, speeds things up
5718 * dbus/dbus-string.c (_dbus_string_free): apply align offset
5719 when freeing the string
5720 (_dbus_string_steal_data): fix for align offset
5721 (undo_alignment): new function
5723 2003-02-26 Havoc Pennington <hp@redhat.com>
5725 All kinds of audit fixes from Owen, plus initial attempt to
5726 handle unaligned memory returned from malloc.
5728 * dbus/dbus-string.c (_dbus_string_init): clamp max length to
5729 leave room for align_offset and nul byte
5730 (fixup_alignment): function to track an align_offset and
5731 ensure real->str is aligned
5732 (DBUS_GENERIC_STRING_PREAMBLE): len must be less than allocated,
5733 to allow a nul byte plus align offset
5734 (_dbus_string_lock): fix overflow issue
5735 (_dbus_string_init_const_len): add assertions on sanity of len,
5736 assign allocated to be ALLOCATION_PADDING larger than len
5737 (set_length): fixup the overflow handling
5738 (_dbus_string_get_data_len): fix overflow in assertion
5739 (open_gap): detect overflow in size of gap to be opened
5740 (_dbus_string_lengthen): add overflow check
5741 (_dbus_string_align_length): fix overflow with _DBUS_ALIGN_VALUE
5742 (_dbus_string_append): add overflow check
5743 (_dbus_string_append_unichar): overflow
5744 (_dbus_string_delete): fix overflow in assertion
5745 (_dbus_string_copy_len): overflow in assertion
5746 (_dbus_string_replace_len): overflows in assertions
5747 (_dbus_string_find): change to implement in terms of
5748 _dbus_string_find_to
5749 (_dbus_string_find_to): assorted fixage
5750 (_dbus_string_equal_c_str): assert c_str != NULL,
5751 fix logic so the function works
5752 (_dbus_string_ends_with_c_str): fix overflow thingy
5753 (_dbus_string_base64_encode): overflow fix
5754 (_dbus_string_validate_ascii): overflow
5755 (_dbus_string_validate_nul): overflow
5757 2003-02-26 Havoc Pennington <hp@redhat.com>
5759 * dbus/dbus-marshal.c (_dbus_marshal_test): fix to work with DISABLE_ASSERTS
5761 2003-02-26 Alexander Larsson <alexl@redhat.com>
5764 Set DBUS_GLIB_THREADS_LIBS for apps using gthread-2.0
5766 * dbus/dbus-connection.c:
5767 * dbus/dbus-connection.h:
5768 Fix _dbus_connection_acquire_io_path and _dbus_connection_acquire_dispatch.
5769 Add dbus_connection_set_wakeup_main_function and use it when queueing
5770 incoming and outgoing messages.
5773 * dbus/dbus-dataslot.c:
5774 Threadsafe usage of DBusDataSlotAllocator
5776 * dbus/dbus-message.c: (dbus_message_get_args_iter):
5779 * dbus/dbus-server-unix.c:
5782 * glib/dbus-gmain.c:
5783 Implement the new wakeup functions for glib.
5786 * glib/test-thread-client.c:
5787 * glib/test-thread-server.c:
5788 * glib/test-thread.h:
5789 Initial cut at some thread test code. Not really done yet.
5791 2003-02-26 Havoc Pennington <hp@pobox.com>
5793 * dbus/dbus-connection.c
5794 (dbus_connection_send_message_with_reply_and_block): fix crash
5795 where we ref'd the outgoing message instead of the returned reply
5797 * dbus/dbus-transport-unix.c (do_authentication): check read watch
5798 at the end of this function, so if we didn't need to read for
5799 authentication, we reinstall it for receiving messages
5801 * dbus/dbus-message.c (dbus_message_new_reply): allow replies to
5802 a NULL sender for peer-to-peer case
5804 * dbus/dbus-transport-unix.c (check_read_watch): handle
5805 !authenticated case correctly
5807 * glib/dbus-gmain.c: add support for DBusServer
5809 * dbus/dbus-server.c: add data slot support
5811 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): check
5812 return values and handle errors
5814 * dbus/dbus-dataslot.c: factor out the data slot stuff from
5817 * Doxyfile.in (INPUT): add glib subdir
5819 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): rename
5820 setup_with_g_main instead of hookup_with_g_main; write docs
5822 2003-02-24 Anders Carlsson <andersca@codefactory.se>
5824 * dbus/dbus-marshal.c: (_dbus_marshal_validate_arg):
5825 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
5826 * dbus/dbus-message.c: (dbus_message_append_boolean),
5827 (dbus_message_append_boolean_array),
5828 (dbus_message_get_args_valist), (_dbus_message_test):
5829 * dbus/dbus-message.h:
5830 * doc/dbus-specification.sgml:
5831 Various fixes as pointed out by Havoc.
5833 * test/data/invalid-messages/bad-boolean-array.message:
5834 * test/data/invalid-messages/bad-boolean.message:
5835 Add invalid boolean value test cases.
5837 2003-02-24 Anders Carlsson <andersca@codefactory.se>
5839 * dbus/dbus-internals.c: (_dbus_type_to_string):
5840 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
5841 (_dbus_marshal_validate_arg):
5842 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
5843 * dbus/dbus-message.c: (dbus_message_append_args_valist),
5844 (dbus_message_append_boolean), (dbus_message_append_boolean_array),
5845 (dbus_message_get_args_valist), (dbus_message_iter_get_boolean),
5846 (dbus_message_iter_get_int32), (dbus_message_iter_get_uint32),
5847 (dbus_message_iter_get_double),
5848 (dbus_message_iter_get_boolean_array), (message_iter_test):
5849 * dbus/dbus-message.h:
5850 * dbus/dbus-protocol.h:
5851 * doc/dbus-specification.sgml:
5852 * test/data/valid-messages/lots-of-arguments.message:
5853 Add support for boolean and boolean array types.
5855 2003-02-23 Havoc Pennington <hp@pobox.com>
5857 * dbus/dbus-keyring.c: finish most of this implementation and
5860 * dbus/dbus-errors.c (dbus_set_error_const, dbus_set_error): make
5861 these barf if the error isn't cleared to NULL
5863 * dbus/dbus-sysdeps.c (_dbus_delete_file): set error on failure
5864 (_dbus_create_directory): new function
5866 * dbus/dbus-errors.c (dbus_set_error): fix warning
5868 * dbus/dbus-string.c (_dbus_string_hex_encode): new function
5869 (_dbus_string_hex_decode): new function
5870 (test_hex_roundtrip): test code
5872 * dbus/dbus-sha.c (_dbus_sha_compute): use dbus_string_hex_encode
5874 * dbus/dbus-md5.c (_dbus_md5_compute): use dbus_string_hex_encode
5876 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): make this use
5877 the save-to-temp/rename trick to atomically write the new file
5878 (_dbus_string_parse_uint): new function
5880 2003-02-22 Havoc Pennington <hp@pobox.com>
5882 * test/Makefile.am (dist-hook): fix dist for test/data/sha-1
5884 2003-02-22 Havoc Pennington <hp@pobox.com>
5886 * dbus/dbus-message.c (dbus_message_iter_get_string_array):
5887 (dbus_message_iter_get_byte_array): Fix up doxygen warnings
5889 * dbus/dbus-sha.c: add implementation of SHA-1 algorithm
5891 * dbus/test/data/sha-1: add US government test suite for SHA-1
5893 2003-02-21 Anders Carlsson <andersca@codefactory.se>
5895 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
5896 Make string arrays NULL-terminated.
5898 * dbus/dbus-memory.c: (dbus_free_string_array):
5899 * dbus/dbus-memory.h:
5900 New function for freeing NULL-terminated string arrays.
5902 * dbus/dbus-message-builder.c: (append_quoted_string),
5903 (_dbus_message_data_load):
5904 Add support for array types.
5906 * dbus/dbus-message.c: (check_message_handling):
5907 Add more types as test cases.
5909 * dbus/dbus-sysdeps.c: (_dbus_string_parse_int),
5910 (_dbus_string_parse_double):
5911 Add the start offset to the end offset.
5913 * test/data/valid-messages/lots-of-arguments.message:
5914 New test message with lots of arguments.
5916 2003-02-21 Anders Carlsson <andersca@codefactory.se>
5918 * dbus/dbus-message.c: (dbus_message_append_nil),
5919 (dbus_message_append_int32), (dbus_message_append_uint32),
5920 (dbus_message_append_double), (dbus_message_append_string),
5921 (dbus_message_append_int32_array),
5922 (dbus_message_append_uint32_array),
5923 (dbus_message_append_double_array),
5924 (dbus_message_append_byte_array),
5925 (dbus_message_append_string_array):
5926 Fix all out-of-memory handling in these functions.
5928 2003-02-21 Anders Carlsson <andersca@codefactory.se>
5930 * dbus/dbus-message.c: (dbus_message_append_nil):
5933 2003-02-21 Anders Carlsson <andersca@codefactory.se>
5935 * dbus/dbus-message.c: (dbus_message_append_args_valist),
5936 (dbus_message_append_nil), (dbus_message_append_int32_array),
5937 (dbus_message_append_uint32_array),
5938 (dbus_message_append_double_array),
5939 (dbus_message_append_byte_array),
5940 (dbus_message_append_string_array), (dbus_message_get_args_valist),
5941 (dbus_message_iter_get_int32_array),
5942 (dbus_message_iter_get_uint32_array),
5943 (dbus_message_iter_get_double_array),
5944 (dbus_message_iter_get_byte_array),
5945 (dbus_message_iter_get_string_array):
5947 * dbus/dbus-message.h:
5948 Add functions for appending and getting arrays.
5950 2003-02-21 Anders Carlsson <andersca@codefactory.se>
5952 * dbus/dbus-mempool.c (_dbus_mem_pool_new): Make the
5953 element size at least 8 bytes, fixes mempool tests on
5956 2003-02-20 Alexander Larsson <alexl@redhat.com>
5958 * dbus/dbus-transport-unix.c (unix_do_iteration):
5959 Unlock the connection mutex during a blocking select call.
5960 Add todo about how we need a way to wake up the select.
5962 * dbus/dbus-connection-internal.h:
5963 * dbus/dbus-connection.c:
5964 Add _dbus_connection_lock and _dbus_connection_unlock.
5966 2003-02-19 Havoc Pennington <hp@pobox.com>
5968 * Doxyfile.in (PREDEFINED): put DOXYGEN_SHOULD_SKIP_THIS in
5969 Doxyfile.in, not Doxyfile
5971 * dbus/dbus-keyring.c: do some hacking on this
5973 * dbus/dbus-sysdeps.c (_dbus_delete_file): new
5975 * dbus/dbus-errors.c (dbus_set_error_const): do not call
5977 (dbus_set_error): remove dbus_error_init, check for message ==
5978 NULL *before* we sprintf into it, and add @todo about including
5979 system headers in this file
5981 * dbus/dbus-sysdeps.c (_dbus_create_file_exclusively): new
5983 * dbus/dbus-errors.h (DBUS_ERROR_FAILED): add
5985 * dbus/dbus-sysdeps.c (get_user_info): break this function out to
5986 get various bits of user information based on either username
5988 (_dbus_homedir_from_username): new function
5990 2003-02-19 Anders Carlsson <andersca@codefactory.se>
5993 Add check for nonposix getpwnam_r
5995 * dbus/dbus-mempool.c: (_dbus_mem_pool_new):
5996 Align the pool element size to a sizeof (void *) boundary.
5998 * dbus/dbus-sysdeps.c: (_dbus_setenv), (_dbus_connect_unix_socket),
5999 (_dbus_listen_unix_socket), (_dbus_credentials_from_username):
6000 General Solaris fixes.
6002 * test/data/valid-messages/simplest-manual.message:
6003 Explicitly state that we want little-endian packing.
6005 2003-02-19 Mikael Hallendal <micke@codefactory.se>
6007 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
6009 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_tcp_socket):
6010 Added to create a transport connecting using a tcp/ip socket.
6012 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): Added to connect
6013 to a tcp socket at given host and port.
6014 (_dbus_listen_tcp_socket): added to listen on tcp socket for given
6017 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
6019 * dbus/dbus-server-unix.c (_dbus_server_new_for_tcp_socket):
6020 Added to create a server listening on a TCP/IP socket.
6022 2003-02-19 Havoc Pennington <hp@pobox.com>
6024 Throughout: mop up all the Doxygen warnings and undocumented
6027 * dbus/dbus-sysdeps.c (do_exec): do not use execvp, we don't want
6028 to search any paths.
6030 * dbus/dbus-threads.c: move global mutex initializers to
6031 dbus-internals.h, multiple prototypes was confusing doxygen
6032 besides being kind of ugly
6034 * Doxyfile (PREDEFINED): have Doxygen define
6035 DOXYGEN_SHOULD_SKIP_THIS so we can exclude things from
6036 docs with #ifndef DOXYGEN_SHOULD_SKIP_THIS
6037 (do not abuse the feature! it's for stuff like the autogenerated
6038 macros in dbus-md5.c, not just for things you don't feel like
6041 2003-02-18 Havoc Pennington <hp@pobox.com>
6043 * dbus/dbus-string.c (_dbus_string_zero): new function
6045 * dbus/dbus-md5.c: include MD5 implementation by L. Peter Deutsch,
6046 wrap it in some dbus-friendly API
6048 * dbus/dbus-types.h: add 16-bit types
6050 2003-02-18 Joe Shaw <joe@assbarn.com>
6052 * dbus/dbus-auth.c (handle_server_data_stupid_test_mech): Just get
6053 credentials from our currently running process.
6054 (get_word): Fix a buglet where we were copying the entire length
6055 instead of relative to our position.
6057 * dbus/dbus-hash.c (_dbus_hash_test): Don't try to allocate the
6058 keys on the stack... it's 640k of data.
6060 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): Always
6061 read the credentials byte off the socket, even if we don't have
6063 (_dbus_poll): Implement poll() using select() for systems which
6066 * glib/test-dbus-glib.c (main): Print out an error if no
6067 parameters are given.
6069 * test/data/auth/fallback.auth-script: Added. Tests that a client
6070 can fallback to a secondary auth mechanism if the first fails.
6072 2003-02-18 Havoc Pennington <hp@pobox.com>
6076 2003-02-17 Havoc Pennington <hp@pobox.com>
6078 * doc/dbus-specification.sgml: lots of cosmetic
6079 cleanups/rearrangement, add assorted FIXME, change DBUS_ADDRESS
6080 env variable to DBUS_BUS_ADDRESS, s/client/application/,
6081 s/server/bus/ (except in authentication section). Add a section
6082 "Message Bus Message Routing"
6084 2003-02-17 Anders Carlsson <andersca@codefactory.se.>
6090 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6092 * doc/dbus-specification.sgml:
6093 Specification updates.
6095 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6097 * bus/activation.c: (bus_activation_init), (child_setup),
6098 (bus_activation_activate_service):
6100 * bus/main.c: (main):
6101 Set DBUS_ADDRESS environment variable.
6103 * dbus/dbus-errors.c: (dbus_set_error):
6104 Don't use va_copy since that's a C99 feature.
6106 * dbus/dbus-sysdeps.c: (_dbus_setenv), (do_exec),
6107 (_dbus_spawn_async):
6108 * dbus/dbus-sysdeps.h:
6109 Add child_setup_func to _dbus_spawn_async.
6111 * doc/dbus-specification.sgml:
6112 Update specification.
6114 * test/spawn-test.c: (setup_func), (main):
6117 2003-02-17 Alexander Larsson <alexl@redhat.com>
6119 * dbus/dbus-connection.c (_dbus_connection_handler_destroyed_locked):
6122 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6126 * doc/dbus-test-plan.sgml:
6127 Add test plan document.
6132 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6134 * dbus/dbus-message.c: (decode_header_data),
6135 (_dbus_message_loader_return_buffer):
6136 Set the header padding amount when loading a message.
6138 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6140 * bus/dispatch.c: (send_one_message):
6141 Only send broadcast messages to registered connections.
6143 * dbus/dbus-message.c: (dbus_message_name_is):
6144 * dbus/dbus-message.h:
6145 New convenience function.
6147 * dbus/dbus-transport-debug.c: (do_reading):
6148 Only dispatch one message per run.
6151 * test/bus-test.c: (new_connection_callback), (die),
6152 (test_hello_client1_handler), (test_hello_client2_handler),
6153 (test_hello_replies), (main):
6155 * test/bus-test-loop.[ch]:
6158 2003-02-16 Havoc Pennington <hp@pobox.com>
6160 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
6161 backward conditional
6163 2003-02-16 Alexander Larsson <alexl@redhat.com>
6165 * dbus/dbus-connection.c:
6166 Implement sent_message_with_reply. (with_reply_and block is still
6168 Made dispatch_message not lose message if OOM.
6170 * dbus/dbus-errors.h:
6171 Add NoReply error (for reply timeouts).
6173 2003-02-16 Alexander Larsson <alexl@redhat.com>
6175 * dbus/dbus-hash.c (_dbus_hash_table_unref):
6176 Actually free keys and values when destroying hashtable.
6178 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6180 * dbus/dbus-auth.c: (client_try_next_mechanism):
6183 * dbus/dbus-threads.c: (dbus_condvar_wait_timeout):
6184 Return TRUE if there's no thread implementation around.
6186 * glib/dbus-gmain.c: (free_source),
6187 (dbus_connection_hookup_with_g_main):
6188 Make sure to remove the GSource when the connection is finalized.
6190 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6192 * bus/dispatch.c: (bus_dispatch_message_handler):
6193 * dbus/dbus-errors.h:
6194 Return an error if someone tries to send a message to a service
6197 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6199 * bus/activation.c: (load_directory), (bus_activation_init),
6200 (bus_activation_activate_service):
6203 (bus_driver_handle_activate_service), (bus_driver_handle_message):
6204 More work on the activation handling.
6206 * dbus/dbus-errors.h:
6207 Add some error messages
6209 * dbus/dbus-message.c: (dbus_message_new_error_reply):
6210 * dbus/dbus-message.h:
6211 New function that creates an error message.
6213 * dbus/dbus-protocol.h:
6214 Add ACTIVATE_SERVER message.
6216 * dbus/dbus-server-unix.c: (unix_handle_watch),
6217 (_dbus_server_new_for_domain_socket):
6218 Call _dbus_fd_set_close_on_exec.
6220 * dbus/dbus-sysdeps.c: (make_pipe), (do_exec),
6221 (_dbus_spawn_async), (_dbus_disable_sigpipe),
6222 (_dbus_fd_set_close_on_exec):
6223 * dbus/dbus-sysdeps.h:
6224 Add _dbus_fd_set_close_on exec function. Also add function that checks
6225 that all open fds are set to close-on-exec and warns otherwise.
6227 * dbus/dbus-transport-unix.c:
6228 (_dbus_transport_new_for_domain_socket):
6229 Call _dbus_fd_set_close_on_exec.
6231 2003-02-16 Havoc Pennington <hp@pobox.com>
6233 * dbus/dbus-connection.c (dbus_connection_set_change_sigpipe):
6234 allow people to avoid setting SIGPIPE to SIG_IGN
6235 (_dbus_connection_new_for_transport): disable SIGPIPE unless
6236 we've been asked not to
6238 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6240 * dbus/dbus-list.c: (_dbus_list_append_link),
6241 (_dbus_list_prepend_link):
6242 * dbus/dbus-memory.c: (dbus_malloc), (dbus_malloc0),
6246 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6249 * bus/activation.c: (bus_activation_entry_free),
6250 (add_desktop_file_entry), (load_directory), (bus_activation_init):
6252 * bus/main.c: (main):
6253 Add simple activation support, doesn't work yet though.
6255 2003-02-15 Zack Rusin <zack@kde.org>
6257 * qt/dbus-qthread.cpp: small casting fix
6259 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6261 * dbus/dbus-errors.c: (dbus_set_error):
6262 * dbus/dbus-errors.h:
6263 Add a few errors and make dbus_set_error void.
6265 * dbus/dbus-sysdeps.c:
6266 (_dbus_errno_to_string), (close_and_invalidate), (make_pipe),
6267 (write_err_and_exit), (read_ints), (do_exec), (_dbus_spawn_async):
6268 * dbus/dbus-sysdeps.h:
6269 Add _dbus_spawn_async.
6271 * test/spawn-test.c: (main):
6272 Test for _dbus_spawn_async.
6274 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6276 * dbus/dbus-internals.h:
6277 Fix build without tests.
6279 * dbus/dbus-list.c: (alloc_link):
6280 Fix a segfault when a malloc fails.
6282 * dbus/dbus-memory.c: (initialize_malloc_debug), (dbus_malloc),
6283 (dbus_malloc0), (dbus_realloc):
6284 Add support for malloc debugging.
6286 2003-02-15 Alexander Larsson <alexl@redhat.com>
6288 * dbus/dbus-threads.c:
6289 * dbus/dbus-threads.h:
6290 Add condvars. Remove static mutext from API.
6291 Implement static mutexes by initializing them from threads_init.
6293 * glib/dbus-gthread.c:
6294 * qt/dbus-qthread.cpp:
6295 Update with the thread api changes.
6300 Turn StaticMutex into normal mutex + init function.
6301 Export new functions _dbus_list_alloc_link, _dbus_list_free_link,
6302 _dbus_list_append_link, _dbus_list_prepend_link
6305 * dbus/dbus-sysdeps.c:
6306 * dbus/dbus-sysdeps.h:
6307 New type dbus_atomic_t, and new functions _dbus_atomic_inc,
6308 _dbus_atomic_dec. Only slow fallback implementation at the moment.
6310 * dbus/dbus-protocol.h:
6311 Add DBUS_MESSAGE_LOCAL_DISCONNECT define
6313 * dbus/dbus-message.c:
6314 Make ref/unref atomic.
6317 * dbus/dbus-connection-internal.h:
6318 * dbus/dbus-connection.c:
6319 * dbus/dbus-connection.h:
6321 Change _peek to _borrow,_return & _steal_borrowed.
6322 Change disconnect callback to event.
6323 Make dbus_connection_dispatch_messages reentrant.
6325 * dbus/dbus-transport.c:
6326 Don't ref the connection on calls to the transport
6329 * dbus/dbus-message-handler.c:
6332 * glib/dbus-gmain.c:
6333 Don't use peek_message anymore
6336 * test/debug-thread.c:
6337 * test/debug-thread.h:
6338 Simple thread implementation that asserts() on deadlocks in
6339 single-threaded code.
6342 (main) Call debug_threads_init.
6345 Use disconnect message instead of disconnect callback.
6349 Don't call dbus_connection_set_disconnect_function. Instead export
6350 bus_connection_disconnect.
6353 Call bus_connection_disconnect when we get a disconnected message.
6355 2003-02-15 Havoc Pennington <hp@pobox.com>
6357 * dbus/dbus-message.c (dbus_message_new): fool around with the
6360 2003-02-14 Havoc Pennington <hp@pobox.com>
6362 * dbus/dbus-mempool.c: fail if the debug functions so indicate
6364 * dbus/dbus-memory.c: fail if the debug functions indicate we
6367 * dbus/dbus-internals.c (_dbus_set_fail_alloc_counter)
6368 (_dbus_decrement_fail_alloc_counter): debug functions to
6369 simulate memory allocation failures
6371 2003-02-14 Havoc Pennington <hp@pobox.com>
6373 * dbus/dbus-errors.h (struct DBusError): add a word of padding
6376 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6378 * bus/driver.c: (bus_driver_handle_hello):
6380 * bus/services.c: (bus_service_lookup):
6381 Reorder message sending so we get a more sane order.
6383 * test/bus-test.c: (message_handler):
6386 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6388 * bus/driver.c: (bus_driver_send_service_deleted),
6389 (bus_driver_send_service_created), (bus_driver_send_service_lost),
6390 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
6391 (bus_driver_send_welcome_message),
6392 (bus_driver_handle_list_services),
6393 (bus_driver_handle_acquire_service),
6394 (bus_driver_handle_service_exists):
6395 * dbus/dbus-bus.c: (dbus_bus_register_client),
6396 (dbus_bus_acquire_service), (dbus_bus_service_exists):
6397 * dbus/dbus-errors.c: (dbus_result_to_string):
6398 * dbus/dbus-errors.h:
6399 * dbus/dbus-message.c: (dbus_message_append_args),
6400 (dbus_message_append_args_valist), (dbus_message_get_args),
6401 (dbus_message_get_args_valist), (dbus_message_get_args_iter),
6402 (dbus_message_iter_get_arg_type), (dbus_message_iter_get_string),
6403 (dbus_message_iter_get_byte_array),
6404 (dbus_message_iter_get_string_array), (message_iter_test),
6405 (check_message_handling), (_dbus_message_test):
6406 * dbus/dbus-message.h:
6407 * test/bus-test.c: (main):
6408 Change fields to arguments in messages, so that they won't be
6409 confused with header fields.
6411 * glib/test-dbus-glib.c: (main):
6412 Remove append_fields from hello message.
6414 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6416 * dbus/dbus-errors.c:
6417 * dbus/dbus-message.c:
6418 * dbus/dbus-string.c:
6419 Documentation fixes.
6421 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6423 * glib/dbus-gmain.c: (timeout_handler), (add_timeout),
6425 Implement support for timeouts in dbus-glib.
6427 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6429 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
6430 * dbus/dbus-message.c: (process_test_subdir):
6431 * test/break-loader.c: (find_breaks_based_on):
6432 Plug some memory leaks.
6434 2003-02-13 Richard Hult <rhult@codefactory.se>
6436 * bus/main.c: Fix build.
6438 * dbus/dbus-errors.h:
6439 * dbus/dbus-errors.c: Fix copyright for Anders.
6441 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6446 * bus/connection.c: (bus_connection_foreach):
6449 * bus/desktop-file.c: (grow_lines_in_section), (grow_sections),
6450 (unescape_string), (new_section), (parse_section_start),
6451 (parse_key_value), (report_error), (bus_desktop_file_load),
6452 (bus_desktop_file_get_string):
6453 * bus/desktop-file.h:
6454 Use DBusError for error reporting.
6456 * bus/dispatch.c: (send_one_message),
6457 (bus_dispatch_message_handler):
6458 * bus/driver.c: (bus_driver_send_service_deleted),
6459 (bus_driver_send_service_created), (bus_driver_send_service_lost),
6460 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
6461 (bus_driver_send_welcome_message),
6462 (bus_driver_handle_list_services),
6463 (bus_driver_handle_acquire_service),
6464 (bus_driver_handle_service_exists):
6465 * bus/loop.c: (bus_loop_run):
6467 Use BUS_HANDLE_OOM instead of _DBUS_HANDLE_OOM.
6469 * bus/utils.c: (bus_wait_for_memory):
6471 New files with general utility functions.
6473 * dbus/dbus-internals.h:
6474 Remove _DBUS_HANDLE_OOM.
6476 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6478 * dbus/dbus-errors.c: (dbus_result_to_string), (dbus_error_init),
6479 (dbus_error_free), (dbus_set_error_const), (dbus_set_error):
6480 * dbus/dbus-errors.h:
6481 Add DBusError structure.
6483 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6485 * test/data/valid-messages/standard-acquire-service.message:
6486 * test/data/valid-messages/standard-hello.message:
6487 * test/data/valid-messages/standard-list-services.message:
6488 * test/data/valid-messages/standard-service-exists.message:
6489 Add some standard messages.
6491 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6493 * bus/driver.c: (bus_driver_send_welcome_message),
6494 (bus_driver_handle_list_services),
6495 (bus_driver_handle_acquire_service),
6496 (bus_driver_handle_service_exists), (bus_driver_handle_message):
6497 Update for API changes in libdbus.
6499 * dbus/dbus-message.c: (dbus_message_new_reply):
6500 * dbus/dbus-message.h:
6501 Remove the name argument. The spec states that replies shouldn't
6504 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6506 * bus/desktop-file.c: (parse_section_start), (parse_key_value),
6507 (report_error), (bus_desktop_file_load), (lookup_section),
6508 (lookup_line), (bus_desktop_file_get_raw),
6509 (bus_desktop_file_get_string):
6510 * bus/desktop-file.h:
6511 Some fixes, and new functions for getting a key value from a section.
6513 2003-02-13 Havoc Pennington <hp@pobox.com>
6515 * test/data/auth/fail-after-n-attempts.auth-script: new test
6517 * dbus/dbus-auth.c (send_rejected): shutdown_mech() when we
6520 2003-02-13 Havoc Pennington <hp@pobox.com>
6522 * dbus/dbus-auth.c (handle_server_data_external_mech): args to
6523 dbus_credentials_match were backward
6525 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support
6526 NO_CREDENTIALS and ROOT_CREDENTIALS
6528 * dbus/dbus-auth.c (_dbus_auth_do_work): move get_state() routine
6529 into here. Never process more commands after we've reached an
6530 end state; store further data as unused bytes.
6532 * test/data/auth/*: add more auth tests
6534 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support EXPECT
6535 command to match exact string and EXPECT_UNUSED to match unused
6538 * test/Makefile.am (dist-hook): fix to dist all the test stuff
6540 2003-02-12 Havoc Pennington <hp@pobox.com>
6542 * dbus/dbus-string.c (_dbus_string_pop_line): fix to also strip
6543 \r off of popped lines
6545 * dbus/dbus-auth.c (_dbus_auth_test): write code to run auth
6548 * dbus/dbus-auth-script.c (_dbus_auth_script_run): when doing a
6551 2003-02-12 Havoc Pennington <hp@pobox.com>
6553 * dbus/Makefile.am: remove break-loader from the build, since it
6556 * configure.in: add --enable-gcov to turn on coverage profiling
6557 flags and disable optimization
6559 2003-02-10 Havoc Pennington <hp@pobox.com>
6561 * dbus/dbus-auth-script.c, dbus/dbus-auth-script.h: sync
6562 initial cut at test framework for DBusAuth from laptop.
6563 Doesn't quite work yet but it compiles and I need to get
6564 it off the 266mhz laptop. ;-)
6566 * dbus/dbus-server-debug.c (_dbus_server_debug_accept_transport):
6567 fix a memleak in error case
6569 2003-02-12 Anders Carlsson <andersca@codefactory.se>
6572 * bus/desktop-file.c:
6573 * bus/desktop-file.h:
6574 Add a desktop file parser.
6576 2003-02-11 Zack Rusin <zack@kde.org>
6578 * qt/message.[h|cpp]: sample implementation
6579 of the KDE wrapper for DBusMessage
6581 2003-02-09 Zack Rusin <zack@kde.org>
6583 * test/bus-test.c: make_it_compile
6584 * doc/dbus-specification.sgml: minimal semantic fix
6586 2003-02-06 Anders Carlsson <andersca@codefactory.se>
6592 2003-02-06 Anders Carlsson <andersca@codefactory.se>
6595 * dbus/dbus-break-loader.c:
6597 * test/break-loader.c:
6598 Move dbus-break-loader to test/ and rename it to break-loader.
6600 2003-02-02 Havoc Pennington <hp@pobox.com>
6602 * dbus/dbus-keyring.c, dbus/dbus-keyring.h: template files
6603 for code to manage cookies in your home directory
6605 * dbus/dbus-sysdeps.c (_dbus_generate_random_bytes): new function
6607 * dbus/dbus-auth.c (get_state): impose a maximum number of tries
6608 to authenticate, then disconnect the client.
6610 2003-02-03 Alexander Larsson <alexl@redhat.com>
6612 * dbus/dbus-message.c (dbus_message_append_fields):
6615 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6617 * doc/dbus-specification.sgml:
6618 Update address format section.
6620 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6623 * test/bus-test.c: (get_time), (add_timeout), (remove_timeout),
6624 (message_handler), (new_connection_callback), (loop_quit),
6628 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6630 * bus/driver.c: (bus_driver_handle_service_exists):
6631 Simplify the code a bit.
6633 * dbus/dbus-bus.c: (dbus_bus_service_exists):
6636 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6639 Add libdbus-daemon.la and link to it.
6641 2003-02-01 James Willcox <jwillcox@gnome.org>
6643 * bus/driver.c: (bus_driver_handle_own_service):
6644 Actually include the service reply code in the message.
6646 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6648 * bus/driver.c: (bus_driver_handle_service_exists):
6649 Don't unref the incoming message.
6651 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6653 * dbus/dbus.h: Add dbus-address.h and dbus-bus.h
6655 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6657 * dbus/dbus-server.c: (dbus_server_listen):
6658 * dbus/dbus-transport.c: (_dbus_transport_open):
6659 ifdef out the calls to the debug transport and server.
6661 2003-02-02 Alexander Larsson <alexl@redhat.com>
6663 * dbus/dbus-watch.c (dbus_watch_get_flags):
6664 Add note in the docs that ERROR or HANGUP won't be returned
6665 and are assumed always on.
6667 * glib/dbus-gmain.c (add_watch):
6668 Always add IO_ERR | IO_HUP
6670 * dbus/dbus-message.h:
6671 Add semicolon after dbus_message_iter_get_string_array().
6672 Makes qt code build again
6674 2003-02-01 Anders Carlsson <andersca@codefactory.se>
6676 * bus/driver.c: (create_unique_client_name),
6677 (bus_driver_handle_hello):
6678 Don't take a name, just use a numeric id to identify
6682 * dbus/dbus-bus.c: (dbus_bus_register_client),
6683 (dbus_bus_acquire_service), (dbus_bus_service_exists):
6685 Add new convenience functions for communicating with the bus.
6687 * dbus/dbus-message.h:
6689 * dbus/dbus-protocol.h:
6692 2003-02-01 Alexander Larsson <alexl@redhat.com>
6694 * dbus/dbus-message.c (dbus_message_append_fields):
6695 Add some more doc comments.
6697 2003-02-01 Havoc Pennington <hp@pobox.com>
6699 * dbus/dbus-break-loader.c (randomly_modify_length): change
6700 a 4-byte value in the message as if it were a length
6702 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): don't set
6703 execute bit on saved files
6705 2003-02-01 Havoc Pennington <hp@pobox.com>
6707 * dbus/dbus-break-loader.c (main): new program to find messages
6708 that break the loader.
6710 * dbus/dbus-sysdeps.c (_dbus_string_append_uint): new function
6711 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): new function
6713 * dbus/dbus-string.c (_dbus_string_set_byte): new
6715 2003-01-31 Havoc Pennington <hp@pobox.com>
6717 * dbus/dbus-message.c: refactor the test code to be more general,
6718 in preparation for writing a "randomly permute test cases to
6719 try to break the loader" program.
6721 2003-01-31 Havoc Pennington <hp@pobox.com>
6723 * doc/dbus-specification.sgml: work on the specification
6725 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): check
6726 the protocol version of the message.
6728 * dbus/dbus-protocol.h: drop special _REPLY names, the spec
6729 no longer specifies that.
6730 (DBUS_SERVICE_REPLY_SERVICE_EXISTS): fix flags (1/2/4/8 not
6733 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos): add missing
6734 "break" for DBUS_TYPE_NIL, remove @todo
6736 2003-01-31 Havoc Pennington <hp@pobox.com>
6738 * dbus/dbus-message.c (dbus_message_set_is_error_reply): rename
6739 just set_is_error/get_is_error as this is a commonly-used
6740 function, and write docs.
6742 2003-01-31 Anders Carlsson <andersca@codefactory.se>
6744 * dbus/dbus-address.c: (dbus_address_entry_free):
6745 Free key and value lists.
6747 * dbus/dbus-internals.c: (_dbus_type_to_string):
6748 Add the types we didn't have.
6750 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
6751 (_dbus_marshal_validate_arg):
6754 * dbus/dbus-message.c: (dbus_message_set_sender):
6755 Remove todo about being able to set sender to NULL.
6757 (dbus_message_set_is_error_reply),
6758 (dbus_message_get_is_error_reply):
6759 * dbus/dbus-message.h:
6762 * dbus/dbus-protocol.h:
6763 Add error reply flag.
6765 * test/data/valid-messages/opposite-endian.message:
6766 Add NIL type to test.
6768 2003-01-31 Havoc Pennington <hp@pobox.com>
6770 * doc/dbus-specification.sgml: fully specify the header. Add
6771 flags and major protocol version, and change header/body len to
6774 * dbus/dbus-message-builder.c (append_saved_length): append length
6777 * dbus/dbus-message.c (dbus_message_create_header): change header
6778 length and body length to unsigned. Add the new fields from the
6780 (_dbus_message_loader_return_buffer): unsigned header/body len
6782 2003-01-30 Havoc Pennington <hp@pobox.com>
6784 * dbus/dbus-auth.c: rework to use only REJECTED, no
6787 * doc/dbus-sasl-profile.txt: drop MECHANISMS and just
6788 use REJECTED, suggested by Mark McLoughlin
6790 2003-01-30 Havoc Pennington <hp@pobox.com>
6792 * dbus/dbus-server.c (dbus_server_listen): @todo about how we need
6793 a better way to report errors here. e.g. "unix address lacks
6794 path" or something. also "no such file" when the path doesn't
6797 * dbus/dbus-address.c (dbus_address_entries_free): add @todo about
6799 (dbus_parse_address): add @todo about documenting address format,
6800 and allowing , and ; to be escaped
6802 2003-01-30 Anders Carlsson <andersca@codefactory.se>
6805 Add dbus-address.[ch]
6807 * dbus/dbus-address.c: (dbus_address_entry_free),
6808 (dbus_address_entries_free), (create_entry),
6809 (dbus_address_entry_get_method), (dbus_address_entry_get_value),
6810 (dbus_parse_address), (_dbus_address_test):
6811 * dbus/dbus-address.h:
6812 New files for dealing with address parsing.
6814 * dbus/dbus-connection.c:
6815 Document timeout functions.
6817 * dbus/dbus-message.c:
6818 Document dbus_message_new_from_message.
6820 * dbus/dbus-server-debug.c:
6823 * dbus/dbus-server.c: (dbus_server_listen):
6824 Parse address and use correct server implementation.
6826 * dbus/dbus-string.c: (_dbus_string_find_to), (_dbus_string_test):
6827 * dbus/dbus-string.h:
6828 New function with test.
6830 * dbus/dbus-test.c: (dbus_internal_symbol_do_not_use_run_tests):
6834 * dbus/dbus-transport-debug.c:
6837 * dbus/dbus-transport.c: (_dbus_transport_open):
6838 Parse address and use correct transport implementation.
6840 2003-01-30 Havoc Pennington <hp@pobox.com>
6842 * dbus/dbus-message.c: use message->byte_order instead of
6843 DBUS_COMPILER_BYTE_ORDER throughout.
6844 (dbus_message_create_header): pad header to align the
6845 start of the body of the message to 8-byte boundary
6847 * dbus/dbus-marshal.h: make all the demarshalers take const
6848 DBusString arguments.
6850 * dbus/dbus-message.c (_dbus_message_loader_return_buffer):
6851 validate message args here, so we don't have to do slow validation
6852 later, and so we catch bad messages as they are incoming. Also add
6853 better checks on header_len and body_len. Also fill in
6856 * dbus/dbus-string.c (_dbus_string_validate_utf8): new (not
6857 implemented properly)
6858 (_dbus_string_validate_nul): new function to check all-nul
6860 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): rename
6861 get_arg_end_pos and remove all validation
6862 (_dbus_marshal_validate_arg): actually do validation here.
6864 2003-01-29 Havoc Pennington <hp@pobox.com>
6866 * dbus/dbus-message.c (check_message_handling): fix assertion
6867 failure on set_client_serial
6869 2003-01-28 Havoc Pennington <hp@pobox.com>
6871 * dbus/dbus-server-debug.c: Add doc section comments
6873 * dbus/dbus-transport-debug.c: add doc section comments
6875 2003-01-28 Havoc Pennington <hp@redhat.com>
6877 * dbus/dbus-string.c (_dbus_string_base64_decode): append bytes in
6878 the reverse order from how I had it
6879 (_dbus_string_base64_encode): reverse encoding order. I was
6880 basically byteswapping everything during encoding.
6882 2003-01-28 Anders Carlsson <andersca@codefactory.se>
6884 * dbus/dbus-connection-internal.h:
6885 * dbus/dbus-connection.c: (_dbus_connection_add_timeout),
6886 (_dbus_connection_remove_timeout):
6887 Add functions for adding and removing timeouts.
6889 * dbus/dbus-message.c: (dbus_message_new_from_message):
6890 Add new function that takes a message and creates an exact
6891 copy of it, but with the refcount set to 1.
6892 (check_message_handling):
6895 * dbus/dbus-server-protected.h:
6896 * dbus/dbus-server.c: (_dbus_server_init_base),
6897 (_dbus_server_finalize_base), (_dbus_server_add_timeout),
6898 (dbus_server_set_timeout_functions):
6899 (_dbus_server_remove_timeout):
6900 New functions so that a server can add and remove timeouts.
6902 (dbus_server_listen):
6903 Add commented out call to dbus_server_debug_new.
6905 * dbus/dbus-timeout.c: (_dbus_timeout_new):
6906 Actually set the handler, doh.
6908 * dbus/dbus-transport.c: (_dbus_transport_open):
6909 Add commented out call to dbus_transport_debug_client_new.
6912 Add dbus-transport-debug.[ch] and dbus-server-debug.[ch]
6914 2003-01-28 Havoc Pennington <hp@pobox.com>
6916 * dbus/dbus-message.c (check_message_handling): function to check
6917 on the loaded message, iterates over it etc.
6919 2003-01-28 Havoc Pennington <hp@pobox.com>
6921 * test/Makefile.am (dist-hook): fix make distdir
6923 * dbus/Makefile.am (TESTS_ENVIRONMENT): fix make check
6925 2003-01-27 Havoc Pennington <hp@pobox.com>
6927 * dbus/dbus-mempool.c (time_for_size): replace printf with
6930 * dbus/dbus-message-builder.c (_dbus_message_data_load): allow
6931 empty lines; fix the SAVE_LENGTH stuff to be
6932 START_LENGTH/END_LENGTH so it actually works; couple other
6935 * test/Makefile.am (dist-hook): add dist-hook for .message files
6937 * dbus/dbus-string.c (DBUS_STRING_COPY_PREAMBLE): source of a copy
6938 can be constant or locked.
6939 (_dbus_string_free): allow freeing a const string as
6942 * dbus/dbus-sysdeps.c (_dbus_concat_dir_and_file): utility
6944 * dbus/dbus-test-main.c (main): take an argument which is the
6945 directory containing test data
6947 * dbus/dbus-message.c (_dbus_message_test): pass a test_data_dir
6948 argument to this and load all the messages in test/data/
6949 checking that they can be loaded or not loaded as appropriate.
6951 2003-01-27 Anders Carlsson <andersca@codefactory.se>
6953 * bus/dispatch.c: (bus_dispatch_message_handler):
6954 Dispatch messages sent to services.
6956 * bus/driver.c: (bus_driver_send_service_deleted),
6957 (bus_driver_send_service_created), (bus_driver_send_service_lost),
6958 (bus_driver_send_service_acquired):
6959 Add helper functions for sending service related messages.
6961 (bus_driver_send_welcome_message):
6962 Send HELLO_REPLY instead of WELCOME.
6964 (bus_driver_handle_list_services):
6965 Send LIST_SERVICES_REPLY instead of SERVICES.
6967 (bus_driver_handle_own_service),
6968 (bus_driver_handle_service_exists):
6969 New message handlers.
6971 (bus_driver_handle_message):
6972 Invoke new message handlers.
6974 (bus_driver_remove_connection):
6975 Don't remove any services here since that's done automatically
6976 by bus_service_remove_owner now.
6979 New function signatures.
6981 * bus/services.c: (bus_service_add_owner):
6982 Send ServiceAcquired message if we're the only primary owner.
6984 (bus_service_remove_owner):
6985 Send ServiceAcquired/ServiceLost messages.
6987 (bus_service_set_prohibit_replacement),
6988 (bus_service_get_prohibit_replacement):
6989 Functions for setting prohibit replacement.
6991 (bus_service_has_owner):
6992 New function that checks if a connection is in the owner queue of
6996 Add new function signatures.
6998 * dbus/dbus-list.c: (_dbus_list_test):
6999 Add tests for _dbus_list_remove_last and traversing the list backwards.
7002 Fix a typo in _dbus_list_get_prev_link, if we're at the first element we can't
7003 go any further, so return NULL then.
7005 * dbus/dbus-protocol.h:
7006 Add new messages, service flags and service replies.
7008 2003-01-26 Havoc Pennington <hp@pobox.com>
7010 * dbus/dbus-message-builder.c: implement, completely untested.
7012 * test/data/*: add data to be used in testing.
7013 ".message" files are our simple loadable text format.
7014 ".message-raw" will be binary dumps of messages.
7016 * dbus/dbus-string.c (_dbus_string_starts_with_c_str): new
7018 2003-01-26 Havoc Pennington <hp@pobox.com>
7020 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): new function
7022 * dbus/dbus-errors.c (dbus_result_to_string): add
7025 * dbus/dbus-message-builder.c: new file, will contain code to load
7026 up messages from files. Not implemented yet.
7028 2003-01-26 Havoc Pennington <hp@pobox.com>
7030 * dbus/dbus-message.c (dbus_message_set_sender): support deleting
7031 the sender by setting to NULL
7033 2003-01-26 Havoc Pennington <hp@pobox.com>
7035 The unit tests pass, but otherwise untested. If it breaks, the
7036 tests should have been better. ;-)
7038 * bus/driver.c (bus_driver_handle_hello): return if we disconnect
7041 * dbus/dbus-message.c: redo everything so we maintain
7042 message->header as the only copy of the various fields.
7043 This avoids the possibility of out-of-memory in some cases,
7044 for example dbus_message_lock() can't run out of memory anymore,
7045 and avoids extra copying. Figured I may as well go ahead and do
7046 this since it was busted for dbus_message_lock to not return
7047 failure on OOM, and dbus_message_write_header was totally
7048 unchecked for OOM. Also fixed some random other bugs.
7050 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): verify
7051 that strings are nul-terminated. Also, end_pos can be equal
7052 to string length just not greater than, I think.
7053 (_dbus_marshal_set_int32): new function
7054 (_dbus_marshal_set_uint32): new function
7055 (_dbus_marshal_set_string): new function
7057 * dbus/dbus-connection.c (_dbus_connection_new_for_transport): fix
7058 a warning, init timeout_list to NULL
7059 (dbus_connection_send_message): don't use uninitialized variable
7062 * dbus/dbus-string.c (_dbus_string_replace_len): new function
7064 2003-01-26 Anders Carlsson <andersca@codefactory.se>
7066 * bus/driver.c: (bus_driver_handle_hello),
7067 (bus_driver_send_welcome_message):
7070 2003-01-26 Anders Carlsson <andersca@codefactory.se>
7072 * dbus/dbus-auth.c: (process_auth), (_dbus_auth_unref):
7073 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
7074 (dbus_connection_unref):
7075 * dbus/dbus-marshal.c: (_dbus_marshal_test):
7076 * dbus/dbus-message.c: (dbus_message_unref),
7079 (dbus_message_get_fields):
7080 Remove debugging printout.
7082 (_dbus_message_loader_return_buffer):
7083 Don't store the header string.
7085 (_dbus_message_test):
7088 2003-01-26 Richard Hult <rhult@codefactory.se>
7090 * glib/dbus-gmain.c (dbus_connection_dispatch): Traverse a copy of
7091 the file descriptor list, since it can change under us.
7093 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7095 * glib/dbus-gmain.c: (dbus_connection_prepare),
7096 (dbus_connection_check), (dbus_connection_dispatch), (add_watch),
7097 (remove_watch), (dbus_connection_hookup_with_g_main):
7098 Rewrite the glib handling to use its own GSource instead of a
7099 GIOChannel so we can catch messages put in the queue while waiting
7102 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7105 * bus/connection.c: (connection_disconnect_handler),
7106 (connection_watch_callback), (bus_connection_setup):
7107 * bus/dispatch.c: (send_one_message),
7108 (bus_dispatch_broadcast_message), (bus_dispatch_message_handler),
7109 (bus_dispatch_add_connection), (bus_dispatch_remove_connection):
7111 * bus/driver.c: (bus_driver_send_service_deleted),
7112 (bus_driver_send_service_created), (bus_driver_handle_hello),
7113 (bus_driver_send_welcome_message),
7114 (bus_driver_handle_list_services), (bus_driver_remove_connection),
7115 (bus_driver_handle_message):
7117 Refactor code, put the message dispatching in its own file. Use
7118 _DBUS_HANDLE_OOM. Also send ServiceDeleted messages when a client
7121 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7123 * dbus/dbus-internals.h:
7124 Add _DBUS_HANDLE_OOM macro, it doesn't do anything currently.
7126 * dbus/dbus-message.c: (dbus_message_get_sender):
7127 * dbus/dbus-message.h:
7128 Implement dbus_message_get_sender.
7130 * dbus/dbus-protocol.h:
7131 Add message and service defines.
7133 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7135 * dbus/dbus-connection.c: (dbus_connection_send_message):
7136 * dbus/dbus-message-internal.h:
7137 * dbus/dbus-message.c: (_dbus_message_get_client_serial),
7138 (dbus_message_write_header):
7139 Remove _dbus_messag_unlock and don't set the client serial on a
7140 message if one already exists.
7142 2003-01-24 Havoc Pennington <hp@pobox.com>
7144 * dbus/dbus-list.c (alloc_link): put a thread lock on the global
7147 * bus/driver.c (bus_driver_handle_list_services): fix a leak
7150 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7152 * dbus/dbus-list.c: (alloc_link), (free_link):
7153 Use a memory pool for the links.
7155 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7157 * bus/connection.c: (bus_connection_foreach):
7159 Add new bus_connection_foreach function.
7161 * bus/driver.c: (send_one_message), (bus_driver_broadcast_message):
7162 Add function that broadcasts a message to all clients.
7164 (bus_driver_send_service_created), (bus_driver_handle_hello),
7165 (bus_driver_send_welcome_message),
7166 (bus_driver_handle_list_services), (bus_driver_message_handler):
7167 Implement functions that take care of listing services, and notifying
7168 clients when new services are created.
7170 * bus/services.c: (bus_services_list):
7172 Add new function that returns an array of strings with the currently
7173 registered services.
7176 * glib/dbus-gmain.c:
7177 Update copyright year.
7179 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7181 * dbus/dbus-connection.c: (dbus_connection_send_message):
7182 Unlock the message in case it was sent earlier.
7184 (dbus_connection_send_message_with_reply_and_block):
7185 Remove the reply message from the list.
7187 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
7188 Set array_len and new_pos correctly.
7190 (_dbus_marshal_test):
7191 Remove debug output.
7193 * dbus/dbus-message-internal.h:
7194 * dbus/dbus-message.c: (_dbus_message_get_reply_serial):
7195 New function that returns the reply serial.
7197 (_dbus_message_unlock):
7198 New function that unlocks a message and resets its header.
7200 (dbus_message_append_string_array),
7201 (dbus_message_get_fields_valist),
7202 (dbus_message_iter_get_field_type),
7203 (dbus_message_iter_get_string_array),
7204 (dbus_message_get_fields),
7205 (dbus_message_append_fields_valist):
7206 Handle string arrays.
7208 (dbus_message_set_sender):
7209 Make this function public since the bus daemon needs it.
7211 (decode_header_data):
7212 Set the reply serial to -1 initially.
7214 * dbus/dbus-message.h:
7215 Add dbus_message_set_sender.
7217 2003-01-24 Havoc Pennington <hp@pobox.com>
7219 * doc/dbus-specification.sgml: add some stuff
7221 2003-01-22 Havoc Pennington <hp@pobox.com>
7223 * doc/dbus-specification.sgml: Start to document the protocol.
7225 2003-01-22 Havoc Pennington <hp@pobox.com>
7227 * dbus/dbus-connection.c
7228 (dbus_connection_send_message_with_reply_and_block): add some @todo
7230 * bus/driver.c (bus_driver_add_connection): add a FIXME about memleak
7232 2003-01-21 Havoc Pennington <hp@pobox.com>
7234 (patch untested because can't compile)
7236 * bus/driver.c (create_unique_client_name): make this function
7237 never recycle client names. Also, caller should initialize
7240 * dbus/dbus-sysdeps.c (_dbus_get_current_time): new function
7242 2003-01-21 Anders Carlsson <andersca@codefactory.se>
7244 * dbus/dbus-marshal.c: (_dbus_marshal_double),
7245 (_dbus_marshal_int32), (_dbus_marshal_uint32),
7246 (_dbus_marshal_int32_array), (_dbus_marshal_uint32_array),
7247 (_dbus_marshal_double_array), (_dbus_marshal_string_array),
7248 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
7249 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
7250 (_dbus_marshal_get_field_end_pos), (_dbus_marshal_test):
7251 * dbus/dbus-marshal.h:
7252 * dbus/dbus-protocol.h:
7253 Add support for marshalling and demarshalling integer, double
7256 2003-01-21 Anders Carlsson <andersca@codefactory.se>
7261 * bus/connection.c: (connection_disconnect_handler):
7262 Remove the connection from the bus driver's list.
7264 (connection_watch_callback): Dispatch messages.
7266 (free_connection_data): Free connection name.
7268 (bus_connection_setup): Add connection to the bus driver's list.
7269 (bus_connection_remove_owned_service):
7270 (bus_connection_set_name), (bus_connection_get_name):
7271 Add functions for setting and getting the connection's name.
7274 Add function headers.
7276 * bus/driver.c: (create_unique_client_name),
7277 (bus_driver_handle_hello_message),
7278 (bus_driver_send_welcome_message), (bus_driver_message_handler),
7279 (bus_driver_add_connection), (bus_driver_remove_connection):
7282 * bus/services.c: (bus_service_free):
7284 New file that handles communication and registreation with the bus
7287 2003-01-21 Anders Carlsson <andersca@codefactory.se>
7289 * dbus/dbus-connection.c: (dbus_connection_send_message):
7290 Add a new client_serial parameter.
7292 (dbus_connection_send_message_with_reply):
7293 Remove a @todo since we've implemented the blocking function.
7295 (dbus_connection_send_message_with_reply_and_block):
7296 New function that sends a message and waits for a reply and
7297 then returns the reply.
7299 * dbus/dbus-connection.h:
7302 * dbus/dbus-errors.c: (dbus_result_to_string):
7303 * dbus/dbus-errors.h:
7304 Add new DBUS_RESULT.
7306 * dbus/dbus-message-internal.h:
7307 * dbus/dbus-message.c: (_dbus_message_get_reply_serial),
7308 (_dbus_message_set_sender), (dbus_message_write_header),
7309 (dbus_message_new_reply), (decode_header_data),
7310 (_dbus_message_loader_return_buffer), (_dbus_message_test):
7311 * dbus/dbus-message.h:
7312 Add new functions that set the reply serial and sender.
7313 Also marshal and demarshal them correctly and add test.
7315 * dbus/dbus-protocol.h:
7316 Add new DBUS_MESSAGE_TYPE_SENDER.
7319 * glib/dbus-gmain.c: (watch_callback), (free_callback_data),
7320 (add_watch), (remove_watch), (add_timeout), (remove_timeout),
7321 (dbus_connection_hookup_with_g_main):
7322 * glib/test-dbus-glib.c: (main):
7323 Rewrite to use GIOChannel and remove the GSource crack.
7325 * test/echo-client.c: (main):
7326 * test/watch.c: (check_messages):
7327 Update for changed APIs
7329 2003-01-19 Anders Carlsson <andersca@codefactory.se>
7331 * dbus/Makefile.am: Add dbus-timeout.[cħ]
7333 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport):
7334 Create a DBusTimeoutList.
7335 (dbus_connection_set_timeout_functions): Add new function to
7336 set timeout callbacks
7338 * dbus/dbus-connection.h: Add public DBusTimeout API.
7340 * dbus/dbus-message.c: (dbus_message_get_service):
7341 * dbus/dbus-message.h: New function.
7343 * dbus/dbus-server.c: Fix small doc typo.
7345 * dbus/dbus-timeout.[ch]: New files for mainloop timeouts.
7347 2003-01-19 Anders Carlsson <andersca@codefactory.se>
7349 * dbus/dbus-string.c (_dbus_string_move_len): Don't delete all
7350 of the string, just as long as specified.
7352 2003-01-19 Havoc Pennington <hp@pobox.com>
7354 * dbus/dbus-connection.c (dbus_connection_get_is_authenticated):
7357 * dbus/dbus-server.c (dbus_server_set_max_connections)
7358 (dbus_server_get_max_connections, dbus_server_get_n_connections):
7359 keep track of current number of connections, and add API for
7360 setting a max (but haven't implemented enforcing the max yet)
7362 2003-01-18 Havoc Pennington <hp@pobox.com>
7364 * dbus/dbus-transport-unix.c (unix_do_iteration): only do the
7365 reading/writing if read_watch != NULL or write_watch != NULL.
7367 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): fix
7368 the message loader code to actually load message->header and
7369 message->body into the newly-created message.
7371 * dbus/dbus-transport-unix.c (check_write_watch): fix a mem leak
7374 * dbus/dbus-connection.c (dbus_connection_set_max_message_size)
7375 (dbus_connection_get_max_message_size)
7376 (dbus_connection_set_max_live_messages_size)
7377 (dbus_connection_get_max_live_messages_size): implement some
7378 resource limitation functions
7380 * dbus/dbus-resources.c: new file implementing some of the
7381 resource limits stuff
7383 * dbus/dbus-message.c (dbus_message_iter_get_byte_array): add
7384 missing docs, add @todo to handle OOM etc.
7386 * dbus/dbus-marshal.c (_dbus_demarshal_byte_array): add missing
7389 2003-01-18 Havoc Pennington <hp@pobox.com>
7391 * dbus/dbus-connection.c (dbus_connection_unref): disconnect the
7392 connection if it hasn't been already.
7394 * dbus/dbus-connection.h: kill off the idea of an ErrorFunction,
7395 replace with DisconnectFunction.
7397 2003-01-18 Havoc Pennington <hp@pobox.com>
7399 Building --disable-verbose-mode --disable-asserts --disable-tests
7400 cuts the library from 112K to 45K or so
7402 * configure.in: check for varargs macro support,
7403 add --enable-verbose-mode, --enable-asserts.
7405 * dbus/dbus-internals.h (_dbus_assert): support
7407 (_dbus_verbose): support DBUS_ENABLE_VERBOSE_MODE
7409 2003-01-18 Havoc Pennington <hp@pobox.com>
7411 * dbus/dbus-test.c: include config.h so that tests actually run
7413 * dbus/dbus-string.c: add assertions that stuff is 8-byte aligned,
7414 so the failure mode when that assumption fails will be plenty
7417 2003-01-18 Havoc Pennington <hp@pobox.com>
7419 * configure.in: default --enable-tests to $USE_MAINTAINER_MODE
7421 * dbus/Makefile.am: fix it up so dubs-test-main.c is included in
7424 * test/Makefile.am: don't use special variable "TESTS" for echo-*
7425 since we don't want to use those in make check
7427 2003-01-15 Havoc Pennington <hp@redhat.com>
7433 2003-01-15 Havoc Pennington <hp@redhat.com>
7435 * test/Makefile.am: fix so that test source code ends up in the
7436 distribution on make distcheck
7438 2003-01-15 Havoc Pennington <hp@redhat.com>
7444 2003-01-15 Havoc Pennington <hp@redhat.com>
7446 * dbus/dbus-test.c (dbus_internal_symbol_do_not_use_run_tests):
7447 fix build when --disable-tests
7449 * Makefile.am (EXTRA_DIST): put HACKING in here
7451 * HACKING: document procedure for making a tarball release.
7453 2003-01-14 Anders Carlsson <andersca@codefactory.se>
7455 * bus/connection.c: (connection_error_handler),
7456 (bus_connection_setup):
7457 * bus/main.c: (main):
7458 Make sure that the DBusConnectionData struct is NULLed
7459 out to prevent a segfault.
7461 * dbus/dbus-errors.c: (dbus_result_to_string):
7462 * dbus/dbus-errors.h:
7463 * dbus/dbus-message.c: (dbus_message_get_fields),
7464 (dbus_message_get_fields_valist), (_dbus_message_test):
7465 * dbus/dbus-message.h:
7466 Make dbus_message_get_fields return a result code so we can
7467 track invalid fields as well as oom.
7469 2003-01-11 Havoc Pennington <hp@pobox.com>
7471 * configure.in: change --enable-test/--enable-ansi action-if-given
7472 to enable_foo=$enableval instead of enable_foo=yes
7474 2003-01-08 Havoc Pennington <hp@pobox.com>
7476 * dbus/dbus-string.c (_dbus_string_align_length): new function
7478 * dbus/dbus-test-main.c: move main() for test app here
7480 (dbus_internal_symbol_do_not_use_run_tests): we have to export a
7481 symbol to run tests, because dbus-test isn't in the main
7484 Code review nitpicks.
7486 * dbus/dbus-message.c (dbus_message_write_header): add newlines
7487 for people with narrow emacs ;-). Assert client_serial was filled
7488 in. Assert message->name != NULL.
7489 (dbus_message_append_fields): have "first_field_type" arg separate
7490 from va list, needed for C++ binding that also uses varargs IIRC
7491 and helps with type safety
7492 (dbus_message_new): add @todo about using DBusString to store
7493 service/name internally
7494 (dbus_message_new): don't leak ->service and ->name on OOM later
7496 (dbus_message_unref): free the service name
7497 (dbus_message_get_fields): same change to varargs
7498 i.e. first_field_type
7499 (_dbus_message_loader_return_buffer): assert that the message data
7500 is aligned (if not it's a bug in our code). Put in verbose griping
7501 about why we set corrupted = TRUE.
7502 (decode_header_data): add FIXME that char* is evil. Was going to
7503 add FIXME about evil locale-specific string.h strncmp, but just
7504 switched to wacky string-as-uint32 optimization. Move check for
7505 "no room for field name" above get_const_data_len() to avoid
7506 assertion failure in get_const_data_len if we have trailing 2
7507 bytes or the like. Check for service and name fields being
7508 provided twice. Don't leak service/name on error. Require field
7509 names to be aligned to 4 bytes.
7511 * dbus/dbus-marshal.c: move byte swap stuff to header
7512 (_dbus_pack_int32): uscore-prefix
7513 (_dbus_unpack_int32): uscore-prefix
7514 (_dbus_unpack_uint32): export
7515 (_dbus_demarshal_string): add @todo complaining about use of
7517 (_dbus_marshal_get_field_end_pos): add @todo about bad error
7518 handling allowing corrupt data to go unchecked
7520 2003-01-08 Havoc Pennington <hp@redhat.com>
7522 * dbus/dbus-transport-unix.c (unix_do_iteration): add read/write
7523 to the select() as needed for authentication. (should be using
7524 _dbus_poll() not select, but for another day)
7526 * dbus/dbus.h: include dbus/dbus-protocol.h
7528 2003-01-08 Anders Carlsson <andersca@codefactory.se>
7530 * dbus/Makefile.am (dbusinclude_HEADERS): Install
7533 2003-01-08 Anders Carlsson <andersca@codefactory.se>
7535 * dbus/dbus-internals.c: (_dbus_type_to_string):
7536 New function that returns a string describing a type.
7538 * dbus/dbus-marshal.c: (_dbus_demarshal_byte_array):
7539 * dbus/dbus-marshal.h:
7540 * dbus/dbus-message.c: (dbus_message_get_fields_valist),
7541 (dbus_message_iter_get_field_type), (dbus_message_iter_get_double),
7542 (dbus_message_iter_get_byte_array):
7543 * dbus/dbus-message.h:
7544 Add new convenience functions for appending and getting message fields.
7545 Also add demarshalling routines for byte arrays.
7547 2003-01-07 Anders Carlsson <andersca@codefactory.se>
7549 * dbus/dbus-connection-internal.h:
7550 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
7551 (_dbus_connection_get_next_client_serial),
7552 (dbus_connection_send_message):
7553 * dbus/dbus-internals.h:
7554 * dbus/dbus-marshal.c: (unpack_uint32), (dbus_unpack_int32),
7555 (dbus_pack_int32), (_dbus_marshal_double), (_dbus_marshal_int32),
7556 (_dbus_marshal_uint32), (_dbus_demarshal_double),
7557 (_dbus_demarshal_int32), (_dbus_demarshal_uint32),
7558 (_dbus_demarshal_string), (_dbus_marshal_get_field_end_pos),
7559 (_dbus_verbose_bytes), (_dbus_marshal_test):
7560 * dbus/dbus-marshal.h:
7561 * dbus/dbus-message-internal.h:
7562 * dbus/dbus-message.c: (_dbus_message_set_client_serial),
7563 (dbus_message_write_header), (_dbus_message_lock),
7564 (dbus_message_new), (dbus_message_ref), (dbus_message_unref),
7565 (dbus_message_get_name), (dbus_message_append_int32),
7566 (dbus_message_append_uint32), (dbus_message_append_double),
7567 (dbus_message_append_string), (dbus_message_append_byte_array),
7568 (dbus_message_get_fields_iter), (dbus_message_iter_ref),
7569 (dbus_message_iter_unref), (dbus_message_iter_has_next),
7570 (dbus_message_iter_next), (dbus_message_iter_get_field_type),
7571 (dbus_message_iter_get_string), (dbus_message_iter_get_int32),
7572 (dbus_message_iter_get_uint32), (dbus_message_iter_get_double),
7573 (decode_header_data), (_dbus_message_loader_return_buffer),
7574 (message_iter_test), (_dbus_message_test):
7575 * dbus/dbus-message.h:
7576 * dbus/dbus-protocol.h:
7577 * dbus/dbus-test.c: (main):
7579 * glib/test-dbus-glib.c: (message_handler), (main):
7580 * test/echo-client.c: (main):
7581 * test/watch.c: (check_messages):
7582 Make messages sendable and receivable for real.
7584 2003-01-07 Anders Carlsson <andersca@codefactory.se>
7586 * dbus/dbus-marshal.c: (_dbus_marshal_double),
7587 (_dbus_marshal_string), (_dbus_marshal_byte_array):
7588 * dbus/dbus-message.c: (dbus_message_append_int32),
7589 (dbus_message_append_uint32), (dbus_message_append_double),
7590 (dbus_message_append_string), (dbus_message_append_byte_array):
7591 Handle OOM restoration.
7593 2003-01-07 Anders Carlsson <andersca@codefactory.se>
7595 * dbus/dbus-marshal.c: (_dbus_marshal_string),
7596 (_dbus_demarshal_string), (_dbus_marshal_test):
7597 * dbus/dbus-marshal.h:
7598 * dbus/dbus-message.c: (dbus_message_get_name),
7599 Document these functions.
7601 (dbus_message_append_int32), (dbus_message_append_uint32),
7602 (dbus_message_append_double), (dbus_message_append_string),
7603 (dbus_message_append_byte_array):
7604 * dbus/dbus-message.h:
7605 Add functions for adding message fields of different types.
7607 * dbus/dbus-protocol.h:
7608 Add the different types.
7610 2003-01-05 Havoc Pennington <hp@pobox.com>
7612 * bus/connection.c: implement routines for handling connections,
7613 first thing is keeping a list of owned services on each connection
7614 and setting up watches etc.
7616 * bus/services.c: implement a mapping from service names to lists
7619 * dbus/dbus-hash.c: add DBUS_HASH_POINTER
7621 * dbus/dbus-threads.c (dbus_static_mutex_lock): add functions
7622 to use static mutexes for global data
7624 * dbus/dbus-connection.c (dbus_connection_set_data): add new
7625 collection of functions to set/get application-specific data
7626 on the DBusConnection.
7628 2003-01-04 Havoc Pennington <hp@pobox.com>
7630 * dbus/dbus-sysdeps.c (_dbus_sleep_milliseconds): new function
7631 (_dbus_poll): new function
7633 * dbus/dbus-internals.h (_DBUS_STRUCT_OFFSET): new macro
7636 * bus/loop.c: initial code for the daemon main loop
7638 2003-01-04 Havoc Pennington <hp@pobox.com>
7640 * test/watch.c (error_handler): make it safe if the error handler
7641 is called multiple times (if we s/error handler/disconnect
7642 handler/ we should just guarantee it's called only once)
7644 * dbus/dbus-transport.c (_dbus_transport_disconnect): call the
7645 error handler on disconnect (it's quite possible we should
7646 just change the error handler to a "disconnect handler," I'm
7647 not sure we have any other meaningful errors)
7649 * configure.in: check for getpwnam_r
7651 * dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
7652 dbus/dbus-auth.c: add credentials support, add EXTERNAL auth
7653 mechanism as in SASL spec, using socket credentials
7655 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): new function
7656 (_dbus_send_credentials_unix_socket): new function
7658 * dbus/dbus-sysdeps.c (_dbus_accept_unix_socket): rename just
7660 (_dbus_write): only check errno if <0 returned
7661 (_dbus_write_two): ditto
7663 2003-01-02 Anders Carlsson <andersca@codefactory.se>
7665 * dbus/dbus-marshal.c: (_dbus_marshal_utf8_string),
7666 (_dbus_marshal_byte_array), (_dbus_demarshal_utf8_string),
7667 (_dbus_marshal_test):
7668 * dbus/dbus-marshal.h:
7669 Add _dbus_marshal_byte_array and rename _dbus_marshal_string
7670 to _dbus_marshal_utf8_string. Also fix some tests.
7672 2002-12-28 Harri Porten <porten@kde.org>
7674 * configure.in: added check for C++ compiler and a very cheesy
7675 check for the Qt integration
7677 * Makefile.am (SUBDIRS): compile qt subdir if support is enabled
7679 * qt/Makefile.am: added
7681 * qt/.cvsignore: added
7683 * qt/dbus-qthread.cc, qt/dbus-qthread.cpp: renamed former to
7684 latter, added #ifdef QT_THREAD_SUPPORT guard.
7686 * dbus/Makefile.am: added missing headers for make dist
7688 2002-12-28 Kristian Rietveld <kris@gtk.org>
7690 * dbus/Makefile.am: fixup export-symbols-regex.
7692 2002-12-27 Anders Carlsson <andersca@codefactory.se>
7694 * acinclude.m4: Add this file and put the
7695 PKG_CHECK_MODULE macro in it.
7697 2002-12-27 Anders Carlsson <andersca@codefactory.se>
7699 * dbus/dbus-marshal.c: (_dbus_marshal_string),
7700 (_dbus_demarshal_double), (_dbus_demarshal_int32),
7701 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
7702 (_dbus_marshal_test):
7703 Make the demarshalling routines align the pos argument.
7704 Add string marshalling tests and fix the obvious bugs
7707 2002-12-26 Havoc Pennington <hp@pobox.com>
7709 * dbus/dbus-auth.c: fixes fixes fixes
7711 * dbus/dbus-transport-unix.c: wire up support for
7712 encoding/decoding data on the wire
7714 * dbus/dbus-auth.c (_dbus_auth_encode_data)
7715 (_dbus_auth_decode_data): append to target string
7716 instead of nuking it.
7718 2002-12-26 Havoc Pennington <hp@pobox.com>
7720 * dbus/dbus-marshal.h (DBUS_COMPILER_BYTE_ORDER): #ifdef
7721 WORDS_BIGENDIAN then compiler byte order is DBUS_BIG_ENDIAN,
7724 * dbus/dbus-marshal.c: Add macros to do int swapping in-place and
7725 avoid swap_bytes() overhead (ignoring possible assembly stuff for
7726 now). Main point is because I wanted unpack_uint32 to implement
7728 (_dbus_verbose_bytes): new function
7730 * dbus/dbus-string.c (_dbus_string_validate_ascii): new function
7732 * dbus/dbus-message.c (_dbus_message_loader_get_is_corrupted): add
7733 mechanism to handle a corrupt message stream
7734 (_dbus_message_loader_new): fix preallocation to only prealloc,
7737 * dbus/dbus-string.c (_dbus_string_skip_blank): fix this function
7738 (_dbus_string_test): enhance tests for copy/move and fix the
7741 * dbus/dbus-transport-unix.c: Hold references in more places to
7742 avoid reentrancy problems
7744 * dbus/dbus-transport.c: ditto
7746 * dbus/dbus-connection.c (dbus_connection_dispatch_message): don't
7747 leak reference count in no-message case
7749 * test/watch.c (do_mainloop): handle adding/removing watches
7750 during iteration over the watches. Also, ref the connection/server
7751 stored on a watch, so we don't try to mangle a destroyed one.
7753 * dbus/dbus-transport-unix.c (do_authentication): perform
7756 * dbus/dbus-auth.c (get_state): add a state
7757 AUTHENTICATED_WITH_UNUSED_BYTES and return it if required
7758 (_dbus_auth_get_unused_bytes): append the unused bytes
7759 to the passed in string, rather than prepend
7761 * dbus/dbus-transport.c (_dbus_transport_init_base): create
7762 the auth conversation DBusAuth
7764 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd)
7765 (_dbus_transport_new_for_domain_socket): when creating a
7766 transport, pass in whether it's a client-side or server-side
7767 transport so we know which DBusAuth to create
7769 2002-12-03 Havoc Pennington <hp@pobox.com>
7771 * dbus/dbus-transport-unix.c (unix_finalize): finalize base
7772 _after_ finalizing the derived members
7773 (unix_connection_set): unref watch if we fail to add it
7775 * dbus/dbus-connection.c (dbus_connection_unref): delete the
7776 transport first, so that the connection owned by the
7777 transport will be valid as the transport finalizes.
7779 * dbus/dbus-transport-unix.c (unix_finalize): free the write_watch
7780 if necessary, and remove watches from the connection.
7782 * dbus/dbus-watch.c (_dbus_watch_list_free): improve a comment
7784 2002-12-26 Anders Carlsson <andersca@codefactory.se>
7786 * dbus/dbus-marshal.c: (_dbus_marshal_string),
7787 (_dbus_demarshal_double), (_dbus_demarshal_int32),
7788 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
7789 (_dbus_marshal_test):
7790 * dbus/dbus-marshal.h:
7791 Add string marshal functions and have the demarshal functions
7792 return the new position.
7794 2002-12-25 Havoc Pennington <hp@pobox.com>
7796 * doc/dbus-sasl-profile.txt: docs on the authentication protocol,
7797 it is a simple protocol that just maps directly to SASL.
7799 * dbus/dbus-auth.h, dbus/dbus-auth.c: authentication protocol
7800 initial implementation, not actually used yet.
7802 * dbus/dbus-string.c (_dbus_string_find): new function
7803 (_dbus_string_equal): new function
7804 (_dbus_string_base64_encode): new function
7805 (_dbus_string_base64_decode): new function
7807 2002-12-25 Anders Carlsson <andersca@codefactory.se>
7810 * dbus/dbus-marshal.c: (swap_bytes), (_dbus_marshal_double),
7811 (_dbus_marshal_int32), (_dbus_marshal_uint32),
7812 (_dbus_demarshal_double), (_dbus_demarshal_int32),
7813 (_dbus_demarshal_uint32), (_dbus_marshal_test):
7814 * dbus/dbus-marshal.h:
7815 * dbus/dbus-protocol.h:
7816 * dbus/dbus-test.c: (main):
7818 Add un-optimized marshalling/demarshalling routines.
7820 2002-12-25 Harri Porten <porten@kde.org>
7822 * qt/dbus-qt.h: adjusted ctor and getter to KDE/Qt conventions
7824 2002-12-24 Zack Rusin <zack@kde.org>
7826 * qt/dbus-qthread.cc: adding - integrates QMutex into Dbus
7827 * qt/dbus-qt.h: skeleton with two sample implemenatation of the
7830 2002-12-24 Havoc Pennington <hp@pobox.com>
7832 * glib/dbus-gthread.c: fix include
7834 * glib/dbus-glib.h: rename DBusMessageHandler for now.
7835 I think glib API needs to change, though, as you don't
7836 want to use DBusMessageFunction, you want to use the
7837 DBusMessageHandler object. Probably
7838 dbus_connection_open_with_g_main_loop()
7839 and dbus_connection_setup_g_main_loop() or something like that
7840 (but think of better names...) that just create a connection
7841 that has watch/timeout functions etc. already set up.
7843 * dbus/dbus-connection.c
7844 (dbus_connection_send_message_with_reply): new function just to
7845 show how the message handler helps us deal with replies.
7847 * dbus/dbus-list.c (_dbus_list_remove_last): new function
7849 * dbus/dbus-string.c (_dbus_string_test): free a string that
7852 * dbus/dbus-hash.c: use memory pools for the hash entries
7853 (rebuild_table): be more paranoid about overflow, and
7854 shrink table when we can
7855 (_dbus_hash_test): reduce number of sprintfs and write
7856 valid C89. Add tests for case where we grow and then
7857 shrink the hash table.
7859 * dbus/dbus-mempool.h, dbus/dbus-mempool.c: memory pools
7861 * dbus/dbus-connection.c (dbus_connection_register_handler)
7862 (dbus_connection_unregister_handler): new functions
7864 * dbus/dbus-message.c (dbus_message_get_name): new
7866 * dbus/dbus-list.c: fix docs typo
7868 * dbus/dbus-message-handler.h, dbus/dbus-message-handler.c:
7869 an object representing a handler for messages.
7871 2002-12-16 Anders Carlsson <andersca@codefactory.se>
7874 * glib/dbus-gthread.c: (dbus_gthread_init):
7875 Don't use the gdbus prefix for public functions.
7877 2002-12-16 Anders Carlsson <andersca@codefactory.se>
7881 Add GLib checks and fixup .pc files
7885 * glib/dbus-gmain.c: (gdbus_connection_prepare),
7886 (gdbus_connection_check), (gdbus_connection_dispatch),
7887 (gdbus_add_connection_watch), (gdbus_remove_connection_watch),
7888 (dbus_connection_gsource_new):
7889 * glib/dbus-gthread.c: (dbus_gmutex_new), (dbus_gmutex_free),
7890 (dbus_gmutex_lock), (dbus_gmutex_unlock), (dbus_gthread_init):
7891 * glib/test-dbus-glib.c: (message_handler), (main):
7894 2002-12-15 Harri Porten <porten@kde.org>
7896 * autogen.sh: check for libtoolize before attempting to use it
7898 * dbus/dbus-transport-unix.c: include <sys/time.h> for timeval
7901 * .cvsignore: ignore more stamp files
7903 * dbus/dbus-watch.c (_dbus_watch_list_new): fixed doc error
7905 * test/Makefile.am: added -I$(top_srcdir) to be able to compile
7906 without make install.
7908 2002-12-15 Havoc Pennington <hp@pobox.com>
7910 * dbus/dbus-threads.c: add thread stubs that a higher library
7911 layer can fill in. e.g. the GLib wrapper might fill them in with
7912 GThread stuff. We still need to use this thread API to
7913 thread-safe-ize the library.
7915 2002-12-12 Havoc Pennington <hp@pobox.com>
7917 * dbus/dbus-transport-unix.c, dbus/dbus-server-unix.c: use the
7918 below new interfaces and include fewer system headers.
7920 * dbus/dbus-sysdeps.c (_dbus_read): new function
7921 (_dbus_write): new function
7922 (_dbus_write_two): new function
7923 (_dbus_connect_unix_socket): new function
7924 (_dbus_listen_unix_socket): new function
7926 * dbus/dbus-message-internal.h: change interfaces to use
7929 2002-12-11 Havoc Pennington <hp@pobox.com>
7931 * dbus/dbus-types.h: add dbus_unichar
7933 * dbus/dbus-internals.c (_dbus_verbose): use _dbus_getenv
7935 * dbus/dbus-connection.c (dbus_connection_send_message): return
7938 * dbus/dbus-transport.c: include dbus-watch.h
7940 * dbus/dbus-connection.c: include dbus-message-internal.h
7942 * HACKING: add file with coding guidelines stuff.
7944 * dbus/dbus-string.h, dbus/dbus-string.c: Encapsulate all string
7945 handling here, for security purposes (as in vsftpd). Not actually
7946 using this class yet.
7948 * dbus/dbus-sysdeps.h, dbus/dbus-sysdeps.c: Encapsulate all
7949 system/libc usage here, as in vsftpd, for ease of auditing (and
7950 should also simplify portability). Haven't actually moved all the
7951 system/libc usage into here yet.
7953 2002-11-25 Havoc Pennington <hp@pobox.com>
7955 * dbus/dbus-internals.c (_dbus_verbose): fix to not
7956 always print the first verbose message.
7958 2002-11-24 Havoc Pennington <hp@pobox.com>
7960 * test/echo-client.c, test/echo-server.c: cheesy test
7963 * configure.in (AC_CHECK_FUNCS): check for writev
7965 * dbus/dbus-message.c (_dbus_message_get_network_data): new
7968 * dbus/dbus-list.c (_dbus_list_foreach): new function
7970 * dbus/dbus-internals.c (_dbus_verbose): new function
7972 * dbus/dbus-server.c, dbus/dbus-server.h: public object
7973 representing a server that listens for connections.
7975 * dbus/.cvsignore: create
7977 * dbus/dbus-errors.h, dbus/dbus-errors.c:
7978 public API for reporting errors
7980 * dbus/dbus-connection.h, dbus/dbus-connection.c:
7981 public object representing a connection that
7982 sends/receives messages. (Same object used for
7983 both client and server.)
7985 * dbus/dbus-transport.h, dbus/dbus-transport.c:
7986 Basic abstraction for different kinds of stream
7987 that we might read/write messages from.
7989 2002-11-23 Havoc Pennington <hp@pobox.com>
7991 * dbus/dbus-internals.h (_DBUS_INT_MAX): add _DBUS_INT_MIN
7994 * dbus/dbus-test.c (main): add list test, and include
7995 dbus-test.h as intended
7997 * dbus/dbus-hash.c (_dbus_hash_table_remove_string)
7998 (_dbus_hash_table_remove_int): return value indicates
7999 whether the entry existed to remove
8001 * dbus/dbus-list.c: add linked list utility class,
8004 * dbus/dbus-hash.c: add TODO item about shrinking the hash bucket
8007 2002-11-23 Havoc Pennington <hp@pobox.com>
8009 * Doxyfile.in (INCLUDE_FILE_PATTERNS): expand DBUS_BEGIN_DECLS/
8010 DBUS_END_DECLS to nothing, that should fix this once and for all
8012 * Doxyfile.in (JAVADOC_AUTOBRIEF): set to YES
8014 * dbus/dbus-message.c, dbus/dbus-hash.c:
8015 add some missing @brief
8017 2002-11-23 Havoc Pennington <hp@pobox.com>
8019 * dbus/dbus-message.h: put semicolons after DEBUG_BEGIN_DECLS
8020 to avoid confusing Doxygen
8022 * dbus/dbus-hash.c: @} not }@
8024 * dbus/dbus-message.c (struct DBusMessage): split out
8027 2002-11-23 Havoc Pennington <hp@pobox.com>
8029 * configure.in: pile on more warning flags if using gcc
8031 * Doxyfile.in (EXTRACT_STATIC): set to NO, so we don't have
8032 to document static functions
8034 * configure.in: add summary to end of configure so it
8035 looks nice and attractive
8037 * dbus/dbus-hash.c: finish implementation and write unit
8040 * configure.in: add --enable-tests to enable unit tests
8042 * dbus/dbus-test.c: test program to run unit tests
8043 for all files in dbus/*, initially runs a test for
8046 * dbus/dbus-internals.h: file to hold some internal utility stuff
8048 2002-11-22 Havoc Pennington <hp@redhat.com>
8050 * dbus/dbus-hash.c: copy in Tcl hash table, not yet
8051 "ported" away from Tcl
8053 * dbus/dbus-types.h: header for types such as dbus_bool_t
8055 2002-11-22 Havoc Pennington <hp@redhat.com>
8057 * dbus/dbus.h: fixups for doc warnings
8059 * Doxyfile.in (FILE_PATTERNS): we need to scan .h to pick up
8061 (QUIET): make it quiet so we can see warnings
8063 * dbus/dbus-memory.c: teach D-BUS to allocate and free memory
8065 2002-11-22 Havoc Pennington <hp@redhat.com>
8067 * Makefile.am: include "Doxyfile" target in all-local
8069 * configure.in: generate the Doxyfile
8071 * Doxyfile.in: move Doxyfile here, so we can use
8072 configure to generate a Doxyfile with the right
8075 2002-11-22 Havoc Pennington <hp@redhat.com>
8077 * dbus/dbus-message.c: move inline docs into .c file
8079 * Doxyfile (OUTPUT_DIRECTORY): move output to doc/api
8080 so all docs are under doc/
8081 (MAN_EXTENSION): generate man pages. Use extension
8082 ".3dbus" which matches ".3qt" on my system,
8083 I guess this is OK, I don't know really.
8084 (FILE_PATTERNS): look for .c files not .h, makes sense
8087 2002-11-22 Havoc Pennington <hp@pobox.com>
8089 * Makefile.am (SUBDIRS): rename subdir "server" to "bus"
8090 because any app can be a server, and any app can be a client,
8091 the bus is a special kind of server.
8093 Thu Nov 21 23:35:31 2002 Zack Rusin <zack@kde.org>
8095 * Doxyfile : adding. Still needs Makefile rules to be generated
8096 automatically (just run "doxygen" in the toplevel dir for now to
8099 * dbus/dbus-message.h : Adding sample docs (javadoc since
8100 resembles gtk-doc a little more)
8102 * dbus/dbus.h : Adding sample docs
8104 2002-11-21 Havoc Pennington <hp@redhat.com>
8106 * dbus/Makefile.am (INCLUDES): define DBUS_COMPILATION
8107 so we can allow ourselves to include files directly,
8108 instead of having to use dbus.h
8110 * dbus/dbus.h: fill in
8112 * dbus/dbus-message.h: sketch out a sample header file.
8113 Include griping if you include it directly instead of
8116 * dbus/dbus-macros.h: new file with macros for extern "C",
8117 TRUE/FALSE, NULL, etc.
8119 * doc/file-boilerplate.c: put include guards in here
8121 2002-11-21 Havoc Pennington <hp@redhat.com>
8123 * doc/file-boilerplate.c: include both AFL and GPL boilerplate.
8125 * COPYING: include the GPL as well, and license code
8126 under both AFL and GPL.
8128 2002-11-21 Havoc Pennington <hp@redhat.com>
8130 * acconfig.h: get rid of this
8132 * autogen.sh (run_configure): add --no-configure option
8134 * configure.in: remove AC_ARG_PROGRAM to make
8135 autoconf complain less. add AC_PREREQ.
8136 add AC_DEFINE third arg.
8138 2002-11-21 Anders Carlsson <andersca@codefactory.se>
8141 Fix references so we can distcheck.
8143 2002-11-21 Havoc Pennington <hp@redhat.com>
8145 * Initial module creation