1 2005-01-21 Havoc Pennington <hp@redhat.com>
3 * dbus/dbus-bus.c: add more return_if_fail checks
5 * dbus/dbus-message.c (load_message): have the "no validation"
6 mode (have to edit the code to toggle the mode for now though)
8 * dbus/dbus-marshal-header.c (_dbus_header_load): have a mode that
9 skips all validation; I want to use this at least for benchmark
10 baseline, I'm not sure if it should be a publicly-available switch.
12 2005-01-21 Havoc Pennington <hp@redhat.com>
14 * glib/dbus-gmain.c: don't put the GLib bindings in the same
15 toplevel doxygen group as the low-level API stuff
17 * dbus/dbus.h: note that libdbus is the low-level API
19 2005-01-20 Havoc Pennington <hp@redhat.com>
21 * update-dbus-docs.sh: script to update docs on the web site, only
22 works for me though. neener.
24 2005-01-20 Havoc Pennington <hp@redhat.com>
26 * dbus/dbus-sysdeps.c (_dbus_poll): amazingly, trying to compile
27 code can reveal bugs in it
29 2005-01-20 Havoc Pennington <hp@redhat.com>
31 * dbus/dbus-sysdeps.c (_dbus_poll): fix several bugs in the
32 select() version, patches from Tor Lillqvist
34 2005-01-20 Havoc Pennington <hp@redhat.com>
36 * doc/dbus-tutorial.xml: replace > with >
38 * bus/services.c (bus_registry_acquire_service): validate the name
39 and return a better error if it's no good.
41 * doc/dbus-specification.xml: note NO_AUTO_START change
43 * dbus/dbus-protocol.h (DBUS_HEADER_FLAG_NO_AUTO_START): change
44 from AUTO_START, we're toggling the default
46 * bus/dispatch.c: adapt the tests to change of auto-start default
48 2005-01-18 Havoc Pennington <hp@redhat.com>
50 * rename dbus-daemon-1 to dbus-daemon throughout
52 2005-01-18 Havoc Pennington <hp@redhat.com>
54 * Throughout, grand renaming to strip out the use of "service",
55 just say "name" instead (or "bus name" when ambiguous). Did not
56 change the internal code of the message bus itself, only the
57 programmer-facing API and messages.
59 * doc/dbus-specification.xml: further update the message bus section
61 * bus/config-parser.c (all_are_equiv): fix bug using freed string
64 2005-01-17 Havoc Pennington <hp@redhat.com>
66 * dbus/dbus-types.h: remove 16-bit types since we don't use them
69 * dbus/dbus-marshal-validate.c (_dbus_validate_path): disallow any
70 "invalid name character" not only non-ASCII
72 * doc/dbus-specification.xml: further update spec, message bus
73 parts are still out-of-date but the marshaling etc. stuff is now
76 2005-01-17 Havoc Pennington <hp@redhat.com>
78 * doc/dbus-specification.xml: partially update spec
80 2005-01-17 Havoc Pennington <hp@redhat.com>
82 * Throughout, align variant bodies according to the contained
83 type, rather than always to 8. Should save a fair bit of space in
86 * dbus/dbus-marshal-validate.c (_dbus_validate_body_with_reason):
87 fix handling of case where p == end
89 * doc/TODO: remove the dbus_bool_t item and variant alignment items
91 2005-01-17 Havoc Pennington <hp@redhat.com>
93 * dbus/dbus-types.h: hardcode dbus_bool_t to 32 bits
95 * Throughout: modify DBUS_TYPE_BOOLEAN to be a 32-bit type instead
96 of an 8-bit type. Now dbus_bool_t is the type to use whenever you
97 are marshaling/unmarshaling a boolean.
99 2005-01-16 Havoc Pennington <hp@redhat.com>
101 This is about it on what can be disabled/deleted from libdbus
102 easily, back below 150K anyhow. Deeper cuts are more work than
103 just turning the code off as I've done here.
105 * dbus/dbus-marshal-basic.c (_dbus_pack_int32): we don't need the
106 signed int convenience funcs
108 * dbus/dbus-internals.c (_dbus_verbose_real): omit when not in
111 * dbus/dbus-string-util.c, dbus/dbus-string.c: more breaking
112 things out of libdbus
114 * dbus/dbus-sysdeps.c, dbus/dbus-sysdeps-util.c: same
116 * dbus/dbus-hash.c: purge the TWO_STRINGS crap (well, make it
117 tests-enabled-only, though it should probably be deleted)
119 * dbus/dbus-message-util.c: same stuff
121 * dbus/dbus-auth-util.c: same stuff
123 2005-01-16 Havoc Pennington <hp@redhat.com>
125 * dbus/dbus-userdb-util.c: split out part of dbus-userdb.c
127 * dbus/dbus-sysdeps.c (_dbus_uid_from_string): move here to pave
128 way for stripping down dbus-userdb.c stuff included in libdbus.
129 Rename _dbus_parse_uid for consistency.
131 2005-01-16 Havoc Pennington <hp@redhat.com>
133 * dbus/dbus-internals.c (_dbus_real_assert): print the function
134 name the assertion failed in
136 * dbus/dbus-internals.h (_dbus_return_if_fail)
137 (_dbus_return_val_if_fail): assert that the name of the function
138 containing the check doesn't start with '_', since we only want to
139 use checks on public functions
141 * dbus/dbus-connection.c (_dbus_connection_ref_unlocked): change
144 * dbus/dbus-marshal-header.c (_dbus_header_set_field_basic):
145 change checks to asserts for private function
147 * dbus/dbus-message.c (_dbus_message_set_serial): checks
148 to asserts for private function
150 * dbus/dbus-marshal-recursive.c (skip_one_complete_type): remove
151 broken assertion that was breaking make check
152 (_dbus_type_reader_array_is_empty): remove this rather than fix
153 it, was only used in assertions
155 2005-01-16 Havoc Pennington <hp@redhat.com>
157 * test/unused-code-gc.py: hacky script to find code that's used
158 only by the bus (not libdbus) or used only by tests or not used at
159 all. It has some false alarms, but looks like we can clean up a
160 lot of size from libdbus.
162 * dbus/dbus-sysdeps.c, dbus/dbus-sysdeps-utils.c,
163 dbus/Makefile.am: initially move 10K of binary size out of libdbus
165 2005-01-16 Havoc Pennington <hp@redhat.com>
167 * Add and fix docs according to Doxygen warnings throughout
170 * dbus/dbus-marshal-recursive.c
171 (_dbus_type_reader_array_is_empty): change this to just call
172 array_reader_get_array_len() and make it static
174 * dbus/dbus-message.c (dbus_message_iter_get_element_type): rename
176 (dbus_message_iter_init_append): rename from append_iter_init
178 * dbus/dbus-marshal-recursive.c
179 (_dbus_type_reader_get_element_type): rename from
180 _dbus_type_reader_get_array_type
182 2005-01-15 Havoc Pennington <hp@redhat.com>
184 * test/glib/test-profile.c (with_bus_server_filter): fix crash
186 * dbus/dbus-marshal-basic.c (_dbus_unpack_uint32): inline as macro
187 when DBUS_DISABLE_ASSERT
188 (_dbus_marshal_set_basic): be sure we align for the string length
190 * dbus/dbus-marshal-recursive.c (skip_one_complete_type): make
193 * dbus/dbus-string.c (_dbus_string_get_const_data_len): add an
195 (_dbus_string_set_byte): provide inline macro version
197 2005-01-15 Havoc Pennington <hp@redhat.com>
199 * Land the new message args API and type system.
201 This patch is huge, but the public API change is not
202 really large. The set of D-BUS types has changed somewhat,
203 and the arg "getters" are more geared toward language bindings;
204 they don't make a copy, etc.
206 There are also some known issues. See these emails for details
208 http://lists.freedesktop.org/archives/dbus/2004-December/001836.html
209 http://lists.freedesktop.org/archives/dbus/2005-January/001922.html
211 * dbus/dbus-marshal-*: all the new stuff
213 * dbus/dbus-message.c: basically rewritten
215 * dbus/dbus-memory.c (check_guards): with "guards" enabled, init
216 freed blocks to be all non-nul bytes so using freed memory is less
219 * dbus/dbus-internals.c (_dbus_test_oom_handling): add
220 DBUS_FAIL_MALLOC=N environment variable, so you can do
221 DBUS_FAIL_MALLOC=0 to skip the out-of-memory checking, or
222 DBUS_FAIL_MALLOC=10 to make it really, really, really slow and
225 * qt/message.cpp: port to the new message args API
226 (operator<<): use str.utf8() rather than str.unicode()
227 (pretty sure this is right from the Qt docs?)
229 * glib/dbus-gvalue.c: port to the new message args API
231 * bus/dispatch.c, bus/driver.c: port to the new message args API
233 * dbus/dbus-string.c (_dbus_string_init_const_len): initialize the
234 "locked" flag to TRUE and align_offset to 0; I guess we never
235 looked at these anyhow, but seems cleaner.
237 * dbus/dbus-string.h (_DBUS_STRING_ALLOCATION_PADDING):
238 move allocation padding macro to this header; use it to implement
239 (_DBUS_STRING_STATIC): ability to declare a static string.
241 * dbus/dbus-message.c (_dbus_message_has_type_interface_member):
242 change to return TRUE if the interface is not set.
244 * dbus/dbus-string.[hc]: move the D-BUS specific validation stuff
245 to dbus-marshal-validate.[hc]
247 * dbus/dbus-marshal-basic.c (_dbus_type_to_string): move here from
250 * dbus/Makefile.am: cut over from dbus-marshal.[hc]
251 to dbus-marshal-*.[hc]
253 * dbus/dbus-object-tree.c (_dbus_decompose_path): move this
254 function here from dbus-marshal.c
256 2005-01-12 Joe Shaw <joeshaw@novell.com>
258 * NEWS: Update for 0.23.
260 * configure.in: Release 0.23.
262 2005-01-12 Joe Shaw <joeshaw@novell.com>
264 * mono/Makefile.am, mono/example/Makefile.am: Always build the
265 dbus DLL with --debug. Clean up after the .mdb files this leaves
268 * mono/doc/Makefile.am: Need to uninstall the docs on "make
271 * mono/Arguments.cs (GetDBusTypeConstructor): If the type
272 is an enum, get the enum's underlying type. Another mono
275 2005-01-11 Joe Shaw <joeshaw@novell.com>
277 Patch from Sjoerd Simons <sjoerd@luon.net>
279 * mono/Makefile.am, mono/example/Makefile.am: Don't redefine
280 DESTDIR. It breaks stuff.
282 2005-01-11 Joe Shaw <joeshaw@novell.com>
284 Patch from Tambet Ingo <tambet@ximian.com>
286 * mono/DBusType/Array.cs (Get): Get the underlying element type by
287 calling type.GetElementType(). The code previously depended on
288 broken Mono behavior, which was fixed in Mono 1.1.3.
290 * mono/DBusType/Dict.cs (constructor): Fix the parameters for
291 Activator.CreateInstance() so that the class's constructor is
292 called with the right parameters.
294 2005-01-11 Joe Shaw <joeshaw@novell.com>
296 Patch from Timo Teräs <ext-timo.teras@nokia.com>
298 * dbus/dbus-connection.c
299 (_dbus_connection_queue_received_message_link): Call
300 _dbus_connection_remove_timeout() instead of the _locked()
301 variant, since it's always called from
302 _dbus_connection_handle_watch(), which handles the locking.
303 Removed the _locked() variant since it's no longer used.
305 2005-01-03 Havoc Pennington <hp@redhat.com>
307 * dbus/dbus-internals.h: I'm an idiot, _dbus_assert certainly can
310 2004-12-26 Havoc Pennington <hp@redhat.com>
312 * dbus/dbus-internals.h: add _DBUS_GNUC_NORETURN to _dbus_assert
314 2005-01-03 Havoc Pennington <hp@redhat.com>
316 * dbus/dbus-sysdeps.c (_dbus_sysdeps_test): fix using == on
319 * dbus/dbus-string.c (_dbus_string_insert_alignment): new function
321 2005-01-02 Havoc Pennington <hp@redhat.com>
323 * dbus/dbus-internals.h (_DBUS_ALIGN_OFFSET): new macro
325 2005-01-01 Havoc Pennington <hp@redhat.com>
327 * configure.in: add -Wfloat-equal
329 2005-01-01 Havoc Pennington <hp@redhat.com>
331 * dbus/dbus-sysdeps.h: add _DBUS_DOUBLES_BITWISE_EQUAL macro,
332 for a variety of reasons '==' doesn't do this.
334 2004-12-31 Havoc Pennington <hp@redhat.com>
336 * dbus/dbus-string.c (_dbus_string_equal_substrings): new function
339 2004-12-30 John (J5) Palmieri <johnp@redhat.com>
341 * python/dbus.py: s/ACTIVATION_REPLY_ACTIVE/ACTIVATION_REPLY_ACTIVATED
343 2004-12-30 John (J5) Palmieri <johnp@redhat.com>
345 * python/dbus_bindings.pyx.in: Change DBUS_ACTIVATION_REPLY_ACTIVATED
346 and DBUS_ACTIVATION_REPLY_ALREADY_ACTIVE to match the values in
347 dbus-protocol.h. Because they are defines and not enums they are not
350 2004-12-26 John (J5) Palmieri <johnp@redhat.com>
352 * python/dbus_bindings.pyx.in (bus_activate_service): Bind
353 dbus_bus_activate_service
355 * python/dbus.py (Bus.activate_service): activate a service on the
358 2004-12-24 Havoc Pennington <hp@redhat.com>
360 * test/decode-gcov.c: change to use .gcno and .gcda files, but the
361 file format has also changed and I haven't adapted to that yet
363 * Makefile.am: load .gcno files from latest gcc
365 2004-12-23 John (J5) Palmieri <johnp@redhat.com>
366 * Patch from Rob Taylor <robtaylor@fastmail.fm>
368 * python/dbus_bindings.pyx.in (bus_get_unix_user): New
371 * python/dbus.py (get_unix_user): Added binding to
372 call dbus_bindings.bus_get_unix_user
374 * python/extract.py: Modified the proto_pat regex to
377 2004-12-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
379 * dbus/make-dbus-glib-error-enum.sh: omit the function keyword for
380 better POSIX compliance.
382 2004-12-19 Havoc Pennington <hp@redhat.com>
384 * dbus/dbus-string.c (_dbus_string_insert_4_aligned)
385 (_dbus_string_insert_8_aligned): new functions
387 * dbus/dbus-string.c (_dbus_string_alloc_space): new function
389 2004-12-18 Havoc Pennington <hp@redhat.com>
391 * dbus/dbus-string.c (_dbus_string_validate_ascii): use ISASCII
394 * dbus/dbus-message.c: fix a comment, and add a still-unused
395 not-implemented function
397 * dbus/dbus-marshal.h: fix comment
399 * dbus/dbus-internals.h (_DBUS_ISASCII): new macro
401 2004-12-17 Joe Shaw <joeshaw@novell.com>
403 * mono/DBusType/Byte.cs, mono/DBusType/Int32.cs,
404 mono/DBusType/Int64.cs, mono/DBusType/UInt32.cs,
405 mono/DBusType/UInt64.cs: Use Enum.GetUnderlyingType() instead of
406 Type.UnderlyingSystemType to get the actual system type
407 underneath. This code previously depended on the broken Mono
408 behavior, which was fixed in 1.1.3.
410 2004-11-27 Havoc Pennington <hp@redhat.com>
412 * dbus/dbus-string.h (_dbus_string_get_byte): inline when asserts
414 (_dbus_string_get_const_data): inline when asserts are disabled
416 * dbus/dbus-message.c: record the _dbus_current_generation of
417 creation so we can complain if dbus_shutdown() is used improperly.
418 Do this only if checks are enabled.
420 * dbus/dbus-connection.c: ditto
422 2004-11-26 Havoc Pennington <hp@redhat.com>
424 * test/glib/test-profile.c: add with_bus mode to profile echoes
425 that go through the bus.
427 * test/glib/run-test.sh: add ability to run test-profile
429 * bus/dbus-daemon-1.1.in: fix to say that SIGHUP causes partial
432 2004-11-26 Havoc Pennington <hp@redhat.com>
434 * test/glib/test-profile.c: clean up how the fake_malloc_overhead
435 thing was implemented
437 2004-11-26 Havoc Pennington <hp@redhat.com>
439 * test/glib/test-profile.c: tweak a bit, add support for some
440 made-up minimal malloc overhead with plain sockets, since in
441 real life some sort of buffers are unavoidable thus we could
442 count them in the theoretical best case
444 2004-11-26 Havoc Pennington <hp@redhat.com>
446 * dbus/dbus-message.c (dbus_message_cache_or_finalize): fix bug
447 where I was trying to cache one too many messages
449 2004-11-26 Havoc Pennington <hp@redhat.com>
451 * dbus/dbus-message.c: reimplement message cache as an array which
452 makes the cache about twice as fast and saves maybe 1.5% overall
454 2004-11-26 Havoc Pennington <hp@redhat.com>
456 * dbus/dbus-threads.c (init_global_locks): forgot to put the
457 message cache lock here
459 2004-11-26 Havoc Pennington <hp@redhat.com>
461 * dbus/dbus-message.c (struct DBusMessage): put the locked bit and
462 the "char byte_order" next to each other to save 4 bytes
463 (dbus_message_new_empty_header): reduce preallocation, since the
464 message cache should achieve a similar effect
465 (dbus_message_cache_or_finalize, dbus_message_get_cached): add a
466 message cache that keeps a few DBusMessage around in a pool,
467 another 8% speedup or so.
469 * dbus/dbus-dataslot.c (_dbus_data_slot_list_clear): new function
471 2004-11-25 Havoc Pennington <hp@redhat.com>
473 * dbus/dbus-transport-unix.c (unix_do_iteration): if we're going
474 to write, without reading or blocking, try it before the poll()
475 and skip the poll() if nothing remains to write. This is about a
476 3% speedup in the echo client/server
478 2004-11-25 Havoc Pennington <hp@redhat.com>
480 The primary change here is to always write() once before adding
481 the write watch, which gives us about a 10% performance increase.
483 * dbus/dbus-transport-unix.c: a number of modifications to cope
484 with removing messages_pending
485 (check_write_watch): properly handle
486 DBUS_AUTH_STATE_WAITING_FOR_MEMORY; adapt to removal of
487 messages_pending stuff
488 (check_read_watch): properly handle WAITING_FOR_MEMORY and
490 (unix_handle_watch): after writing, see if the write watch can be
492 (unix_do_iteration): assert that write_watch/read_watch are
493 non-NULL rather than testing that they aren't, since they
494 aren't allowed to be NULL. check_write_watch() at the end so
495 we add the watch if we did not finish writing (e.g. got EAGAIN)
497 * dbus/dbus-transport-protected.h: remove messages_pending call,
498 since it resulted in too much inefficient watch adding/removing;
499 instead we now require that the transport user does an iteration
500 after queueing outgoing messages, and after trying the first
501 write() we add a write watch if we got EAGAIN or exceeded our
502 max bytes to write per iteration setting
504 * dbus/dbus-string.c (_dbus_string_validate_signature): add this
507 * dbus/dbus-server-unix.c (unix_finalize): the socket name was
508 freed and then accessed, valgrind flagged this bug, fix it
510 * dbus/dbus-message.c: fix several bugs where HEADER_FIELD_LAST was taken
511 as the last valid field plus 1, where really it is equal to the
512 last valid field. Corrects some message corruption issues.
514 * dbus/dbus-mainloop.c: verbosity changes
516 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): handle OOM
517 instead of aborting in one of the test codepaths
519 * dbus/dbus-internals.c (_dbus_verbose_real): fix a bug that
520 caused not printing the pid ever again if a verbose was missing
521 the newline at the end
522 (_dbus_header_field_to_string): add HEADER_FIELD_SIGNATURE
524 * dbus/dbus-connection.c: verbosity changes;
525 (dbus_connection_has_messages_to_send): new function
526 (_dbus_connection_message_sent): no longer call transport->messages_pending
527 (_dbus_connection_send_preallocated_unlocked): do one iteration to
528 try to write() immediately, so we can avoid the write watch. This
529 is the core purpose of this patchset
530 (_dbus_connection_get_dispatch_status_unlocked): if disconnected,
531 dump the outgoing message queue, so nobody will get confused
532 trying to send them or thinking stuff is pending to be sent
534 * bus/test.c: verbosity changes
536 * bus/driver.c: verbosity/assertion changes
538 * bus/dispatch.c: a bunch of little tweaks to get it working again
539 because this patchset changes when/where you need to block.
541 2004-11-23 Havoc Pennington <hp@redhat.com>
543 * test/glib/test-profile.c: modify to accept a plain_sockets
544 argument in which case it will bench plain sockets instead of
545 libdbus, for comparison purposes.
547 2004-11-22 Havoc Pennington <hp@redhat.com>
549 * test/glib/test-profile.c (N_CLIENT_THREADS): run multiple
550 threads for more time, so sysprof can get a grip on it.
552 * dbus/dbus-string.c (_dbus_string_validate_utf8): remove
555 2004-11-13 Havoc Pennington <hp@redhat.com>
557 * test/glib/test-profile.c: fix this thing up a bit
559 * dbus/dbus-message.c (dbus_message_new_empty_header): increase
560 preallocation sizes by a fair bit; not sure if this will be an
561 overall performance win or not, but it does reduce reallocs.
563 * dbus/dbus-string.c (set_length, reallocate_for_length): ignore
564 the test hack that forced constant realloc if asserts are
565 disabled, so we can profile sanely. Sprinkle in some
566 _DBUS_UNLIKELY() which are probably pointless, but before I
567 noticed the real performance problem I put them in.
568 (_dbus_string_validate_utf8): micro-optimize this thing a little
569 bit, though callgrind says it didn't help; then special-case
570 ascii, which did help a lot; then be sure we detect nul bytes as
571 invalid, which is a bugfix.
572 (align_length_then_lengthen): add some more _DBUS_UNLIKELY
573 superstition; use memset to nul the padding instead of a manual
575 (_dbus_string_get_length): inline this as a
576 macro; it showed up in the profile because it's used for loop
579 2004-11-10 Colin Walters <walters@verbum.org>
581 * dbus/dbus-spawn.c (check_babysit_events): Handle EINTR,
584 2004-11-09 Colin Walters <walters@verbum.org>
586 * dbus/dbus-string.c (_dbus_string_get_length): New
587 function, writes DBusString to C buffer.
589 * dbus/dbus-string.h: Prototype it.
591 * dbus/dbus-message.c (dbus_message_type_to_string): New
592 function, converts message type into C string.
594 * dbus/dbus-message.h: Prototype it.
596 * bus/selinux.c (bus_selinux_check): Take source pid,
597 target pid, and audit data. Pass audit data to
599 (log_audit_callback): New function, appends extra
601 (bus_selinux_allows_acquire_service): Also take
602 service name, add it to audit data.
603 (bus_selinux_allows_send): Also take message
604 type, interface, method member, error name,
605 and destination, and add them to audit data.
606 (log_cb): Initialize func_audit.
608 * bus/selinux.h (bus_selinux_allows_acquire_service)
609 (bus_selinux_allows_send): Update prototypes
611 * bus/services.c (bus_registry_acquire_service): Pass
612 service name to bus_selinux_allows_acquire_service.
614 * bus/bus.c (bus_context_check_security_policy): Pass
615 additional audit data. Move assignment of dest
618 2004-11-07 Colin Walters <walters@verbum.org>
620 * dbus/dbus-transport-unix.c (do_authentication): Always
621 initialize auth_completed.
623 2004-11-07 Colin Walters <walters@verbum.org>
625 * bus/bus.c (load_config): Break into three
626 separate functions: process_config_first_time_only,
627 process_config_every_time, and process_config_postinit.
628 (process_config_every_time): Move call of
629 bus_registry_set_service_context_table into
630 process_config_postinit.
631 (process_config_postinit): New function, does
632 any processing that needs to happen late
633 in initialization (and also on reload).
634 (bus_context_new): Instead of calling load_config,
635 open config parser here and call process_config_first_time_only
636 and process_config_every_time directly. Later, after
637 we have forked but before changing UID,
638 invoke bus_selinux_full_init, and then call
639 process_config_postinit.
640 (bus_context_reload_config): As in bus_context_new,
641 load parse file inside here, and call process_config_every_time
642 and process_config_postinit.
644 * bus/services.h, bus/services.c
645 (bus_registry_set_service_context_table): Rename
646 from bus_registry_set_sid_table. Take string hash from config
647 parser, and convert them here into SIDs.
649 * bus/config-parser.c (struct BusConfigParser): Have
650 config parser only store a mapping of service->context
652 (merge_service_context_hash): New function.
653 (merge_included): Merge context string hashes instead
654 of using bus_selinux_id_table_union.
655 (bus_config_parser_new): Don't use bus_selinux_id_table_new;
656 simply create a new string hash.
657 (bus_config_parser_unref): Unref it.
658 (start_selinux_child): Simply insert strings into hash,
659 don't call bus_selinux_id_table_copy_over.
661 * bus/selinux.h, bus/selinux.c (bus_selinux_id_table_union)
662 (bus_selinux_id_table_copy_over): Delete.
664 2004-11-03 Colin Walters <walters@verbum.org>
666 * bus/selinux.c (bus_selinux_pre_init): Kill some unused
669 2004-11-03 Colin Walters <walters@verbum.org>
671 * bus/test-main.c (test_pre_hook): Fix test logic,
672 thanks Joerg Barfurth <Joerg.Barfurth@Sun.COM>.
674 2004-11-02 Colin Walters <walters@redhat.com>
676 * bus/selinux.c (bus_selinux_init): Split into two functions,
677 bus_selinux_pre_init and bus_selinux_post_init.
678 (bus_selinux_pre_init): Just determine whether SELinux is
680 (bus_selinux_post_init): Do everything else.
682 * bus/main.c (main): Call bus_selinux_pre_init before parsing
683 config file, and bus_selinux_post_init after. This ensures that
684 we don't lose the policyreload notification thread that
685 bus_selinux_init created before forking previously.
687 * bus/test-main.c (test_pre_hook): Update for split.
689 2004-10-31 Owen Fraser-Green <owen@discobabe.net>
691 Patch from Johan Fischer <linux@fischaz.com>
693 * mono/doc/Makefile.am (install-data-local): Added directory
696 2004-10-29 Colin Walters <walters@redhat.com>
698 * dbus/dbus-sysdeps.h (_dbus_become_daemon): Also take
699 parameter for fd to write pid to.
701 * dbus/dbus-sysdeps.c (_dbus_become_daemon): Implement it.
703 * bus/bus.c (bus_context_new): Pass print_pid_fd
704 to _dbus_become_daemon (bug #1720)
706 2004-10-29 Colin Walters <walters@redhat.com>
708 Patch from Ed Catmur <ed@catmur.co.uk>
710 * mono/doc/Makefile.am (install-data-local): Handle
713 2004-10-29 Colin Walters <walters@redhat.com>
715 * bus/.cvsignore, qt/.cvsignore: Update.
717 2004-10-29 Colin Walters <walters@redhat.com>
719 Patch from Kristof Vansant <de_lupus@pandora.be>
721 * configure.in: Detect Slackware.
722 * bus/Makefile.am (SCRIPT_IN_FILES): Add rc.messagebus.in.
723 * bus/rc.messagebus.in: New file.
725 2004-10-29 Colin Walters <walters@redhat.com>
727 * tools/dbus-monitor.c (filter_func): Return
728 DBUS_HANDLER_RESULT_HANDLED in filter function
730 http://freedesktop.org/pipermail/dbus/2004-August/001433.html
732 2004-10-29 Colin Walters <walters@redhat.com>
734 Patch from Matthew Rickard <mjricka@epoch.ncsc.mil>
736 * bus/services.c (bus_registry_acquire_service):
737 Correctly retrieve service name from DBusString
740 2004-10-29 Colin Walters <walters@redhat.com>
742 * dbus/dbus-glib.h: Update documentation to not
743 refer to internal APIs.
745 2004-10-27 Joe Shaw <joeshaw@novell.com>
747 * mono/Arguments.cs (GetDBusTypeConstructor):
748 type.UnderlyingSystemType will return "System.Byte" if you do it
749 on "byte[]", which is not what we want. So check the type.IsArray
750 property and use System.Array instead.
752 2004-10-25 John (J5) Palmieri <johnp@redhat.com>
754 * dbus/dbus-sysdeps.c (fill_user_info): On errors do not free
755 the DBusUserInfo structure since this is passed into the function.
756 This would cause a double free when the function that allocated
757 the structure would try to free it when an error occured.
759 * (bus/session.conf.in, bus/Makefile.am, dbus/configure.in):
760 use /usr/share/dbus-1/services instead of /usr/lib/dbus-1.0/services
761 for service activation to avoid 32bit/64bit parallel install issues
763 2004-10-21 Colin Walters <walters@verbum.org>
765 * AUTHORS: Fix my email address, the @gnu.org one
766 has been bouncing for some time. Also add J5.
768 2004-10-21 Colin Walters <walters@verbum.org>
770 * dbus/dbus-transport-unix.c (do_authentication): Return
771 authentication status to callers.
772 (unix_handle_watch): If we completed authentication this round,
773 don't do another read. Instead wait until the next iteration,
774 after we've read any pending data in the auth buffer.
775 (unix_do_iteration): Ditto.
776 (unix_handle_watch): Updated for new do_authentication prototype.
778 2004-10-18 Colin Walters <walters@verbum.org>
780 * bus/selinux.c (bus_selinux_enabled): Handle
781 --disable-selinux case.
783 2004-10-18 Colin Walters <walters@verbum.org>
785 * bus/selinux.h: Add bus_selinux_enabled.
787 * bus/selinux.c (bus_selinux_enabled): Implement it.
789 * bus/config-parser.c (struct include): Add
790 if_selinux_enabled member.
791 (start_busconfig_child): Parse if_selinux_enabled
792 attribute for include.
793 (bus_config_parser_content): Handle it.
795 * bus/session.conf.in, bus/system.conf.in: Add
796 inclusion of context mapping to default config files;
797 conditional on SELinux being enabled.
799 * doc/busconfig.dtd: Add to if_selinux_enabled to default DTD.
801 * test/data/invalid-config-files/badselinux-1.conf,
802 test/data/invalid-config-files/badselinux-2.conf:
803 Test files for bad syntax.
805 2004-10-17 Colin Walters <walters@verbum.org>
807 * dbus/dbus-memory.c (_dbus_initialize_malloc_debug, check_guards)
808 (dbus_malloc, dbus_malloc0, dbus_realloc): Fix up printf
809 format specifier mismatches.
811 2004-10-07 Olivier Andrieu <oliv__a@users.sourceforge.net>
813 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): fix an incorrect
816 * glib/dbus-dbus-gmain.c (dbus_g_bus_get): do not mangle NULL
817 pointer (bug #1540, Leonardo Boiko).
819 2004-09-28 Jon Trowbridge <trow@ximian.com>
821 * mono/BusDriver.cs: Changed BusDriver struct to remove
822 the ServiceCreated and ServiceDeleted events and replace them
823 with the new ServiceOwnerChanged event.
825 * mono/example/BusListener.cs: Added a new example program,
826 which listens for and reports any ServiceOwnerChanged events
829 * mono/example/Makefile.am (DESTDIR): Build changes for the
830 new BusListener.cs example.
832 2004-09-27 Olivier Andrieu <oliv__a@users.sourceforge.net>
834 * bus/signals.c (bus_match_rule_parse): validate the components of
835 match rules (bug #1439).
837 * dbus/dbus-bus.c (dbus_bus_add_match): add a missing OOM test.
839 2004-09-24 Olivier Andrieu <oliv__a@users.sourceforge.net>
841 * doc/dbus-specification.xml: document ServiceOwnerChanged
844 * bus/driver.c, bus/driver.h, bus/services.c: Use
845 ServiceOwnerChanged signal instead of ServiceCreated and
848 * bus/dispatch.c: update testcase for the new signal.
850 2004-09-20 Jon Trowbridge <trow@ximian.com>
852 Patch from Nat Friedman <nat@novell.com>
854 * mono/Makefile.am: A number of small build fixes to allow "make
855 distcheck" to succeed.
857 * mono/example/Makefile.am: "make distcheck" fixes.
859 * mono/AssemblyInfo.cs.in: When signing the assembly, look for the
862 * test/Makefile.am: "make distcheck" fixes.
864 2004-09-17 Olivier Andrieu <oliv__a@users.sourceforge.net>
866 * dbus/dbus-sysdeps.c (_dbus_user_at_console): fix memleak in OOM.
868 * doc/busconfig.dtd: update the DTD for the at_console attribute.
870 * bus/driver.c (bus_driver_handle_hello): correctly handle Hello
871 messages after the first one (bug #1389).
873 * bus/dispatch.c (check_double_hello_message): add a test case for
874 the double hello message bug.
875 (check_existent_service_activation): fix check of spawning error.
877 2004-09-16 David Zeuthen <david@fubar.dk>
879 * python/dbus_bindings.pyx.in: Add support for int64 and uint64
881 2004-09-12 David Zeuthen <david@fubar.dk>
883 Patch from Kay Sievers <kay.sievers@vrfy.org>
885 * bus/bus.c (bus_context_new):
889 Add commandline option --nofork to override configuration file
892 2004-09-09 Olivier Andrieu <oliv__a@users.sourceforge.net>
894 * dbus/dbus-*.h: remove the ; after DBUS_(BEGIN|END)_DECLS. Some C
895 compilers don't like it (bug #974).
897 2004-09-04 Harald Fernengel <harry@kdevelop.org>
899 * qt/connection.*: Applied patch by Jérôme Lodewyck
900 <lodewyck@clipper.ens.fr> to integrate an existing
901 connection into the Qt eventloop
903 2004-08-30 Jon Trowbridge <trow@ximian.com>
905 * mono/BusDriver.cs: Added. This is a class for interacting with
906 the org.freedesktop.DBus service.
908 * mono/Message.cs: Added a mechanism to expose the message that is
909 currently being dispatched via the static Message.Current
910 property. Added Message.Sender and Message.Destination
913 * mono/Handler.cs: Expose the dispatched message via
914 Message.Current when handling method calls.
916 * mono/Service.cs: Expose the dispatched message via
917 Message.Current when handling signal emissions.
919 * mono/Connection.cs: Bind dbus_bus_get_base_service via the
920 Connection.BaseService property.
922 2004-08-28 Havoc Pennington <hp@redhat.com>
924 * dbus/dbus-userdb.c (_dbus_is_console_user): remove unused variable
926 More fixes from Steve Grubb
928 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): fix fd leak
929 (_dbus_listen_tcp_socket): fix fd leak
931 * dbus/dbus-spawn.c (read_pid, read_ints): move the "again:" for
932 EINTR to a bit lower in the code
934 2004-08-26 Jon Trowbridge <trow@ximian.com>
936 * bus/driver.c (bus_driver_handle_service_exists): Respond with
937 TRUE if we are inquiring about the existence of the built-in
938 org.freedesktop.DBus service.
940 2004-08-25 John Palmieri <johnp@redhat.com>
941 * bus/config-parser.c:
942 (struct PolicyType): Add POLICY_CONSOLE
943 (struct Element.d.policy): s/gid_or_uid/gid_uid_or_at_console
944 (start_busconfig_child): Sets up console element when
945 <policy at_console=""> is encountered in a policy file
946 (append_rule_from_element): Convert console elements to console
950 (bus_policy_create_client_policy): Add console rules to the client
951 policy based on if the client is at the console
952 (bus_policy_append_console_rule): New function for adding a
953 console rule to a policy
954 (bus_policy_merge): Handle console rule merging
956 * dbus/dbus-sysdeps.h: Added the DBUS_CONSOLE_DIR constant
957 where we check for console user files
959 * dbus/dbus-sysdeps.c:
960 (_dbus_file_exists): New function which checks if the given
962 (_dbus_user_at_console): New function which does the system
963 specific process of checking if the user is at the console
965 * dbus/dbus-userdb.c:
966 (_dbus_is_console_user): New function converts a UID to user name
967 and then calls the system specific _dbus_user_at_console to
968 see if the user is at the console and therefor a console user
970 2004-08-25 Olivier Andrieu <oliv__a@users.sourceforge.net>
972 * bus/config-parser.c (set_limit):
973 * bus/dbus-daemon-1.1.in:
974 * test/data/valid-config-files/many-rules.conf: set the
975 max_match_rules_per_connection limt from the config file.
977 * doc/busconfig.dtd: update the DTD.
979 * bus/driver.c: remove some unused variables.
981 2004-08-24 Mikael Hallendal <micke@imendio.com>
983 * dbus/dbus-glib-lowlevel.h: Removed dbus_bus_get_with_g_main since
984 it's been replaced by dbus_g_bus_get
986 2004-08-23 Colin Walters <walters@redhat.com>
988 Updated SELinux support from Matthew Rickard <mjricka@epoch.ncsc.mil>
990 * bus/selinux.h: Prototype bus_selinux_get_policy_root.
992 * bus/selinux.c: Create a thread for policy reload notification.
993 (bus_selinux_get_policy_root): Implement.
995 * bus/config-parser.c (start_busconfig_child)
996 (bus_config_parser_content): Support SELinux-root relative
999 * configure.in <HAVE_SELINUX>: Add -lpthread.
1001 * bus/test-main.c (test_pre_hook, test_post_hook): New.
1002 (test_post_hook): Move memory checking into here.
1003 (test_pre_hook, test_post_hook): Move SELinux checks in
1004 here, but conditional on a DBUS_TEST_SELINUX environment
1005 variable. Unfortunately we can't run the SELinux checks
1006 as a normal user, since they won't have any permissions
1007 for /selinux. So this will have to be tested manually
1008 for now, until we have virtualization for most of
1011 2004-08-23 Havoc Pennington <hp@redhat.com>
1013 * dbus/dbus-sysdeps.c (_dbus_change_identity): add setgroups() to
1014 drop supplementary groups, suggested by Steve Grubb
1016 2004-08-20 Colin Walters <walters@redhat.com>
1018 * bus/config-parser.c (start_busconfig_child): Remove some unused
1021 * bus/selinux.c (bus_selinux_id_table_insert): Avoid compiler
1024 2004-08-17 Joe Shaw <joeshaw@novell.com>
1026 * configure.in: If --enable-mono is passed in, if we can't find
1029 * mono/Makefile.am: Use /gacutil to install assemblies into the
1032 2004-08-12 Havoc Pennington <hp@redhat.com>
1034 * NEWS: update for 0.22
1036 * configure.in: release 0.22
1038 2004-08-11 Colin Walters <walters@redhat.com>
1040 * tools/dbus-send.c (main, usage): Add --reply-timeout
1043 2004-08-10 Olivier Andrieu <oliv__a@users.sourceforge.net>
1045 * bus/bus.c (process_config_first_time_only): get rid of an unused
1046 DBusError that was causing a memoy leak (bug #989).
1048 * dbus/dbus-keyring.c, dbus/dbus-message.c:
1049 fix compilation on Solaris/Forte C (bug #974)
1051 * bus/main.c (main): plug two minuscule memleaks.
1053 2004-08-10 Havoc Pennington <hp@redhat.com>
1055 * doc/dbus-tutorial.xml: add some more info on GLib bindings
1057 2004-08-09 Havoc Pennington <hp@redhat.com>
1059 * COPYING: switch to Academic Free License version 2.1 instead of
1060 2.0, to resolve complaints about patent termination clause.
1062 2004-07-31 John (J5) Palmieri <johnp@redhat.com>
1064 * README: added documentation for the --enable-python
1067 2004-07-31 Olivier Andrieu <oliv__a@users.sourceforge.net>
1069 * bus/config-parser.c (bus_config_parser_new): fix an invalid
1070 _unref in the SELinux support.
1072 * doc/busconfig.dtd: update DTD for SELinux support.
1074 * bus/config-loader-libxml.c: fix error handler and parser
1075 initialisation/cleanup. OOM test now works with libxml2 HEAD.
1077 * configure.in: remove the warning about libxml2.
1079 * dbus/dbus-bus.c: silence doxygen warning.
1081 2004-07-31 Colin Walters <walters@redhat.com>
1083 * configure.in: Move #error in SELinux check to its own line.
1085 2004-07-31 Olivier Andrieu <oliv__a@users.sourceforge.net>
1087 * dbus/dbus-internals.h (_DBUS_SET_OOM):
1088 * bus/utils.h (BUS_SET_OOM): use dbus_error_set_const instead of
1091 * bus/dispatch.c (check_send_exit_to_service): fix the test case,
1092 broken by the change in the _SET_OOM macros.
1094 2004-07-31 Colin Walters <walters@redhat.com>
1096 * bus/selinux.c <HAVE_SELINUX>: Include utils.h to get
1099 2004-07-31 Colin Walters <walters@redhat.com>
1101 * configure.in: Use AC_TRY_COMPILE instead of AC_EGREP_HEADER
1102 to correctly detect DBUS__ACQUIRE_SVC. Also add an
1105 2004-07-24 Havoc Pennington <hp@redhat.com>
1107 SELinux support from Matthew Rickard <mjricka@epoch.ncsc.mil>
1109 * bus/selinux.c, bus/selinux.h: new file encapsulating selinux
1112 * configure.in: add --enable-selinux
1114 * bus/policy.c (bus_policy_merge): add FIXME to a comment
1116 * bus/main.c (main): initialize and shut down selinux
1118 * bus/connection.c: store SELinux ID on each connection, to avoid
1119 repeated getting of the string context and converting it into
1122 * bus/bus.c (bus_context_get_policy): new accessor, though it
1124 (bus_context_check_security_policy): check whether the security
1125 context of sender connection can send to the security context of
1126 recipient connection
1128 * bus/config-parser.c: add parsing for <selinux> and <associate>
1130 * dbus/dbus-transport.c (_dbus_transport_get_unix_fd): to
1131 implement dbus_connection_get_unix_fd()
1133 * dbus/dbus-connection.c (dbus_connection_get_unix_fd): new
1134 function, used by the selinux stuff
1136 2004-07-29 Olivier Andrieu <oliv__a@users.sourceforge.net>
1138 * bus/config-loader-libxml.c: complete the implementation of
1139 libxml backend for config file loader. Doesn't work with full OOM
1142 * configure.in: change error when selecting libxml into a warning.
1144 * test/data/invalid-config-files: add two non-well-formed XML
1147 * glib/Makefile.am: libdbus_gtool always uses expat, not libxml.
1149 * dbus/dbus-transport-unix.c (unix_handle_watch): do not
1150 disconnect in case of DBUS_WATCH_HANGUP, several do_reading() may
1151 be necessary to read all the buffer. (bug #894)
1153 * bus/activation.c (bus_activation_activate_service): fix a
1154 potential assertion failure (bug #896). Small optimization in the
1155 case of auto-activation messages.
1157 * dbus/dbus-message.c (verify_test_message, _dbus_message_test):
1158 add test case for byte-through-vararg bug (#901). patch by Kimmo
1161 2004-07-28 Anders Carlsson <andersca@gnome.org>
1164 * python/dbus_bindings.pyx.in:
1165 Add dbus.init_gthreads (), allow emit_signal to pass
1166 arguments to the signal.
1168 2004-07-24 Havoc Pennington <hp@redhat.com>
1170 * AUTHORS: add some people, not really comprehensively, let me
1171 know if I missed you
1173 2004-07-24 Havoc Pennington <hp@redhat.com>
1175 * Makefile.am (DIST_SUBDIRS): add DIST_SUBDIRS, problem solved by
1178 * test/Makefile.am (DIST_SUBDIRS): here also
1180 2004-07-22 Olivier Andrieu <oliv__a@users.sourceforge.net>
1182 * dbus/dbus-sysdeps.c (fill_user_info): fix inexistent label name,
1183 breaking build on Solaris, reported by Farhad Saberi on the ML.
1185 * dbus/dbus-message.c (dbus_message_append_args_valist): fix the
1186 va_arg invocation to account for integer promotion in the case of
1187 DBUS_TYPE_BYTE (unsigned char is promoted to int). (bug #901)
1189 * bus/services.c (bus_service_remove_owner): fix bug #902, use
1190 _dbus_list_get_first_link, not _dbus_list_get_first.
1192 * dbus/dbus-bus.c (dbus_bus_service_exists): plug a memory leak.
1194 * dbus/dbus-object-tree.c (free_subtree_recurse): always null
1195 handler functions so that the asserts in _dbus_object_subtree_unref
1198 * dbus/dbus-transport-unix.c (do_reading):
1199 _dbus_transport_queue_messages return value is of type
1200 dbus_bool_t, not DBusDispatchStatus.
1202 2004-07-19 David Zeuthen <david@fubar.dk>
1204 * dbus/dbus-protocol.h: Add DBUS_ERROR_UNIX_PROCESS_ID_UNKNOWN
1207 (check_get_connection_unix_user): Debug says GetProperty; but the
1208 method is called GetConnectionUnixUser
1209 (check_get_connection_unix_process_id): New function
1210 (bus_dispatch_test): Actually call check_get_connection_unix_user();
1211 also call check_get_connection_unix_process_id()
1214 (bus_driver_handle_get_connection_unix_process_id): New function,
1215 handles GetConnectionUnixProcessID on the org.freedesktop.DBus
1219 (handle_server_data_external_mech): Set pid from the credentials
1220 obtained from the socket
1222 * dbus/dbus-connection.c:
1223 (dbus_connection_get_unix_process_id): New function
1225 * dbus/dbus-connection.h:
1226 Add prototype for dbus_connection_get_unix_process_id
1228 * dbus/dbus-transport.c:
1229 (_dbus_transport_get_unix_process_id): New function
1231 * dbus/dbus-transport.h:
1232 Add prototype for _dbus_transport_get_unix_process_id
1234 2004-07-19 Olivier Andrieu <oliv__a@users.sourceforge.net>
1236 * dbus/dbus-message.c: Message counter fix, patch by Christian
1237 Hammond <chipx86@gnupdate.org>
1239 2004-07-18 Seth Nickell <seth@gnome.org>
1242 * python/dbus_bindings.pyx.in:
1243 * python/tests/test-client.py:
1245 Add dbus.ByteArray and dbus_bindings.ByteArray
1246 types so that byte streams can be passed back.
1248 Give jdahlin the heaps of credit that are so
1251 2004-07-12 Seth Nickell <seth@gnome.org>
1255 Add message argument to the default object_method_handler
1258 * python/dbus_bindings.pyx.in:
1260 Automatically return NIL when passed an empty list
1261 (we can't pass back a list since lists are typed
1262 and we don't have any idea what type the the client
1263 intended the list to be... :-( )
1265 2004-07-10 Seth Nickell <seth@gnome.org>
1267 * python/examples/Makefile.am:
1269 Fix distcheck breakage caused by new examples.
1271 2004-07-10 Seth Nickell <seth@gnome.org>
1275 Add "message" argument to service-side dbus.Object
1276 methods. This will break existing services written
1277 using the python bindings, but will allow extraction
1278 of all the message information (e.g. who its from).
1280 Add improved "object oriented" signal handling/emission.
1282 * python/examples/example-service.py:
1286 * python/examples/example-signal-emitter.py:
1287 * python/examples/example-signal-recipient.py:
1289 Two new examples that show how to emit and receive
1290 signals using the new APIs.
1292 * python/examples/example-signals.py:
1293 * python/examples/gconf-proxy-service.py:
1294 * python/examples/gconf-proxy-service2.py:
1296 Add "message" argument to service methods.
1298 2004-06-28 Kay Sievers <kay.sievers@vrfy.org>
1300 * bus/driver.c (bus_driver_handle_get_connection_unix_user)
1301 * dbus/bus.c (dbus_bus_get_unix_user)
1302 * doc/dbus-specification.xml: implement GetConnectionUnixUser
1303 method of org.freedesktop.DBus interface.
1305 * bus/dispatch.c: test case
1307 2004-06-23 John (J5) Palmieri <johnp@redhat.com>
1309 * python/Makefile.am: switched include directory from glib/ to dbus/
1310 since dbus-glib.h moved
1312 2004-06-22 Olivier Andrieu <oliv__a@users.sourceforge.net>
1314 * configure.in: prevent building the gcj stuff and libxml loader
1315 since they are broken.
1317 2004-06-20 Havoc Pennington <hp@redhat.com>
1319 * dbus/dbus-glib-error-enum.h: autogenerate the GError enum
1320 codes from the dbus error names
1322 * glib/dbus-glib.h: move to subdir dbus/ since it's included
1323 as dbus/dbus-glib.h and that breakage is now visible due to
1324 including dbus/dbus-glib.h in dbus-glib-lowlevel.h
1326 * glib/dbus-glib.h: s/gproxy/g_proxy/
1328 * dbus/dbus-shared.h: new header to hold stuff shared with
1331 * dbus/dbus-protocol.h (DBUS_ERROR_*): move errors here rather
1334 * glib/dbus-glib.h (dbus_set_g_error): move to
1335 dbus-glib-lowlevel.h
1337 * glib/dbus-glib.h: remove dbus/dbus.h from here; change a bunch
1338 of stuff to enable this
1340 * dbus/dbus-glib-lowlevel.h: put dbus/dbus.h here
1342 * a bunch of other changes with the same basic "separate glib
1343 bindings from dbus.h" theme
1345 2004-06-10 Owen Fraser-Green <owen@discobabe.net>
1347 * dbus-sharp.pc.in: Removed glib-sharp inclusion in Libs.
1349 * python/examples/Makefile.am: Fixed typo in EXTRA_DIST.
1351 2004-06-09 Olivier Andrieu <oliv__a@users.sourceforge.net>
1353 * bus/driver.c, dbus/dbus-bus.c: use BOOLEAN instead of UINT32 for
1354 the reply value of the ServiceExists message.
1356 2004-06-07 John (J5) Palmieri <johnp@redhat.com>
1358 * python/dbus_bindings.pyx.in: No longer need to parse path
1359 elements and pass them as arrays of strings. The C API now
1360 accepts plain path strings.
1361 (_build_parsed_path): removed
1363 2004-06-07 Havoc Pennington <hp@redhat.com>
1365 * doc/TODO: remove auto-activation item since it's done; sort
1366 items by importance/milestone
1368 2004-06-07 Havoc Pennington <hp@redhat.com>
1370 * dbus/dbus-message-builder.c (_dbus_message_data_load): append
1371 random signature when using REQUIRED_FIELDS (this hack won't work
1374 * dbus/dbus-message.c: change the signature to be a header field,
1375 instead of message->signature special-case string. Incremental
1376 step forward. Then we can fix up code to send the signature in the
1377 message, then fix up code to validate said signature, then fix up
1378 code to not put the typecodes inline, etc.
1379 (load_one_message): don't make up the signature after the fact
1380 (decode_header_data): require signature field for the known
1383 * dbus/dbus-marshal.c (_dbus_marshal_string_len): new
1385 * dbus/dbus-protocol.h: add DBUS_HEADER_FIELD_SIGNATURE
1387 2004-06-07 Owen Fraser-Green <owen@discobabe.net>
1389 * mono/DBusType/ObjectPath.cs: Renamed PathName argument to Path
1391 * mono/Handler.cs: Updated to follow new path argument for
1392 (un-)registering objects.
1394 * mono/example/Makefile.am:
1396 * configure.in: Bumped required version for mono and use new -pkg
1399 2004-06-05 Olivier Andrieu <oliv__a@users.sourceforge.net>
1401 * dbus/dbus-connection.h, dbus/dbus-connection.c: have object path
1402 registration functions take the path argument as char* instead of
1405 * dbus/dbus-marshal.h, dbus/dbus-marshal.c (_dbus_decompose_path):
1406 split off the path decompostion part of
1407 _dbus_demarshal_object_path. Some misc. fixes to silence compiler
1410 * glib/dbus-gobject.c, test/test-service.c: update accordingly.
1412 2004-06-02 Kristian Høgsberg <krh@redhat.com>
1414 * dbus/dbus-auth.c: Rewrite auth protocol handling to use a state
1415 machine approach. A state is implemented as a function that
1416 handles incoming events as specified for that state.
1418 * doc/dbus-specification.xml: Update auth protocol state machine
1419 specification to match implementation. Remove some leftover
1422 2004-06-02 Kristian Høgsberg <krh@redhat.com>
1424 * glib/dbus-gproxy.c, glib/dbus-gmain.c, dbus/dbus-string.c,
1425 dbus/dbus-object-tree.c, dbus/dbus-message.c: add comments to
1428 * Doxyfile.in: remove deprecated options.
1430 * dbus/dbus-message-handler.c, dbus/dbus-message-handler.h,
1431 glib/test-thread.h, glib/test-thread-client.c,
1432 glib/test-thread-server.c, glib/test-profile.c,
1433 glib/test-dbus-glib.c: remove these unused files.
1435 2004-06-01 Olivier Andrieu <oliv__a@users.sourceforge.net>
1437 * dbus/dbus-object-tree.c
1438 (_dbus_object_tree_dispatch_and_unlock): fix dispatch for
1439 non-fallback handlers (bug #684).
1440 (_dbus_object_subtree_new): initialize invoke_as_fallback field.
1441 (find_subtree_recurse): report wether the returned subtree is an
1442 exact match or a "fallback" match higher up in the tree.
1443 (object_tree_test_iteration): update test case.
1445 2004-06-01 Seth Nickell <seth@gnome.org>
1447 * python/dbus_bindings.pyx.in:
1448 * python/tests/test-client.py:
1450 Round off basic type support. Add dicts (yay!), and
1451 remaining array types.
1453 Make MessageIter more general so it works for dicts too.
1455 Mark all loop variables as C integers.
1457 2004-05-31 Havoc Pennington <hp@redhat.com>
1459 * glib/dbus-gidl.c (method_info_add_arg): keep args sorted with
1462 * glib/dbus-gobject.c (dbus_type_to_string): move to dbus-gutils.c
1464 * glib/dbus-glib-tool.c (main): set up to have a --self-test
1465 option that runs the tests, and start filling in some code
1466 including for starters just dumping the interfaces to stdout
1468 * glib/Makefile.am (INCLUDES): define DBUS_LOCALEDIR
1470 * test/data/valid-introspection-files/lots-of-types.xml: test of
1471 an example introspection file
1473 * glib/dbus-gparser.c (parser_check_doctype): doctype should be
1476 2004-05-31 Seth Nickell <seth@gnome.org>
1478 * python/dbus_bindings.pyx.in:
1479 * python/tests/test-client.py:
1484 Fix string array memory trashing bug... oops...
1486 2004-05-30 Seth Nickell <seth@gnome.org>
1490 Add a nicer-but-less-flexible alternate API for handling
1491 calls to virtual objects in dbus.ObjectTree.
1493 Screw up the argument order to the dbus.Object constructor
1494 for consistency with dbus.ObjectTree (and to make dbus_methods
1495 optional for future extension)
1497 * python/examples/Makefile.am:
1498 * python/examples/gconf-proxy-service.py:
1499 * python/examples/gconf-proxy-service2.py:
1501 Alternate implementation of gconf-proxy-service using the
1502 nicer dbus.ObjectTree API.
1504 * python/examples/example-service.py:
1505 * python/tests/test-server.py
1507 Reverse the argument order to deal with dbus.Object constructor
1510 2004-05-30 Seth Nickell <seth@gnome.org>
1512 * python/examples/example-client.py:
1513 * python/examples/example-service.py:
1515 Take it back. Lists seem to work but they're broken
1516 in the test suite. Make the base examples use
1519 2004-05-30 Seth Nickell <seth@gnome.org>
1521 * python/dbus_bindings.pyx.in:
1522 * python/tests/test-client.py:
1524 Add some more tests and fix errors that crop up.
1525 Unfortunately, currently it seems like marshalling
1526 and unmarshalling of lists is completely broken :-(
1528 2004-05-30 Seth Nickell <seth@gnome.org>
1530 * python/dbus_bindings.pyx.in:
1532 Add support for ObjectPath type.
1536 Refactor message handling code to a common function.
1538 * python/tests/test-client.py:
1539 * python/tests/test-server.py:
1541 Add tests that check to make sure values of all types
1542 can be echoed from a service w/o mangling.
1544 2004-05-29 Seth Nickell <seth@gnome.org>
1548 Add ObjectTree class which allows implementation
1549 of trees of "virtual" objects. Basically the python
1550 wrapper for "register_fallback".
1552 * python/examples/Makefile.am
1553 * python/examples/gconf-proxy-client.py:
1554 * python/examples/gconf-proxy-service.py:
1556 Implement a simple GConf proxy service that supports
1557 get/set on string and int GConf keys using the ObjectTree.
1559 2004-05-29 Seth Nickell <seth@gnome.org>
1562 * python/examples/example-client.py:
1563 * python/examples/example-service.py:
1564 * python/examples/list-system-services.py:
1566 Add SessionBus, SystemBus and ActivationBus classes
1567 so you don't need to know the special little BUS_TYPE
1570 2004-05-29 Havoc Pennington <hp@redhat.com>
1572 * bus/config-parser.c (process_test_valid_subdir): temporarily
1573 stop testing config parser OOM handling, since expat has issues
1574 http://freedesktop.org/pipermail/dbus/2004-May/001153.html
1576 * bus/dbus-daemon-1.1.in: change requested_reply to
1577 send_requested_reply/receive_requested_reply so we can send the
1578 replies, not just receive them.
1580 * bus/config-parser.c: parse the new
1581 send_requested_reply/receive_requested_reply
1583 * bus/policy.c (bus_client_policy_check_can_send): add
1584 requested_reply argument and use it
1586 * bus/bus.c (bus_context_check_security_policy): pass through
1587 requested_reply status to message send check
1589 * bus/system.conf.in: adapt to requested_reply change
1591 2004-05-28 Havoc Pennington <hp@redhat.com>
1593 * test/glib/test-service-glib.c (main): remove unused variable
1595 * glib/dbus-gidl.c (base_info_ref): fix a silly compiler warning
1597 * dbus/dbus-auth.h (enum): remove AUTHENTICATED_WITH_UNUSED_BYTES
1598 from the enum, no longer in use.
1600 * dbus/dbus-sysdeps.h: include config.h so DBUS_VA_COPY actually
1603 * dbus/dbus-message.c: add various _dbus_return_val_if_fail for
1604 whether error_name passed in is a valid error name.
1606 2004-05-28 John (J5) Palmieri <johnp@redhat.com>
1608 * dbus/dbus-message.c (dbus_message_get_args): Added support for
1609 OBJECT_PATH and OBJECT_PATH_ARRAY
1611 2004-05-28 Seth Nickell <seth@gnome.org>
1613 * python/examples/Makefile.am:
1615 Forget to add Makefile.am. Do not pass go.
1617 2004-05-28 Michael Meeks <michael@ximian.com>
1619 * glib/dbus-gvalue.c (dbus_gvalue_marshal, dbus_gvalue_demarshal):
1622 * dbus/dbus-string.c (_dbus_string_parse_basic_type): impl.
1624 * dbus/dbus-message.c (_dbus_message_iter_get_basic_type),
1625 (_dbus_message_iter_get_basic_type_array): impl.
1626 drastically simplify ~all relevant _get methods to use these.
1627 (_dbus_message_iter_append_basic_array),
1628 (dbus_message_iter_append_basic): impl
1629 drastically simplify ~all relevant _append methods to use these.
1631 * dbus/dbus-message-builder.c (parse_basic_type)
1632 (parse_basic_array, lookup_basic_type): impl.
1633 (_dbus_message_data_load): prune scads of duplicate /
1636 * dbus/dbus-marshal.c (_dbus_demarshal_basic_type_array)
1637 (_dbus_demarshal_basic_type): implement,
1638 (demarshal_and_validate_len/arg): beef up debug.
1639 (_dbus_marshal_basic_type, _dbus_marshal_basic_type_array): impl.
1641 2004-05-27 Seth Nickell <seth@gnome.org>
1644 * python/Makefile.am:
1646 Include the example python apps in the tarball.
1648 * python/examples/list-system-services.py
1650 Add a python new example that fetches the list of services
1651 from the system bus.
1653 2004-05-27 Seth Nickell <seth@gnome.org>
1656 * python/dbus_bindings.pyx.in:
1658 Fix failure to notify that a signal was not handled,
1659 resulted in hung functions.
1661 2004-05-25 Colin Walters <walters@redhat.com>
1663 * tools/dbus-monitor.c (main): Monitor all types of messages.
1665 2004-05-23 Owen Fraser-Green <owen@discobabe.net>
1667 * mono/Handler.cs, mono/Service.cs: Added UnregisterObject method
1668 which unregisters the object path and disposes the handler.
1670 2004-05-23 Kristian Høgsberg <krh@redhat.com>
1672 Patch from Timo Teräs <ext-timo.teras@nokia.com> (#614):
1674 * dbus/dbus-message.c (dbus_message_iter_get_args_valist): Swap
1675 operands to && so we call dbus_message_iter_next () for the last
1678 2004-05-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
1680 * dbus/dbus-object-tree.c
1681 (_dbus_object_tree_list_registered_unlock, lookup_subtree): return
1682 children even if the requested path isn't registered.
1683 (object_tree_test_iteration): test object_tree_list_registered.
1685 * configure.in: undefine HAVE_ABSTRACT_SOCKETS instead of defining
1688 2004-05-20 Kristian Høgsberg <krh@redhat.com>
1690 * doc/TODO: Remove resolved items.
1692 * bus/expirelist.h (struct BusExpireList): remove unused n_items
1695 * bus/connection.c (bus_connections_expect_reply): Enforce the
1696 per-connection limit on pending replies.
1698 Patch from Jon Trowbridge <trow@ximian.com>:
1700 * bus/main.c (setup_reload_pipe): Added. Creates a pipe and sets
1701 up a watch that triggers a config reload when one end of the pipe
1703 (signal_handler): Instead of doing the config reload in our SIGHUP
1704 handler, just write to the reload pipe and let the associated
1705 watch handle the reload when control returns to the main loop.
1707 * bus/driver.c (bus_driver_handle_reload_config): Added.
1708 Implements a ReloadConfig method for requesting a configuration
1709 file reload via the bus driver.
1711 2004-05-19 Owen Fraser-Green <owen@discobabe.net>
1713 * HACKING: Updated release instructions concerning the wiki page.
1715 2004-05-18 Kristian Høgsberg <krh@redhat.com>
1717 * dbus/dbus-auth.c (client_try_next_mechanism): Remove logic to
1718 filter against auth->allowed_mechs; we only add allowed mechs in
1719 record_mechanisms().
1721 * dbus/dbus-auth-script.c (_dbus_auth_script_run): Add an
1722 ALLOWED_MECHS to auth-script format so we can set the list of
1725 * data/auth/client-out-of-mechanisms.auth-script: New test to
1726 check client disconnects when it is out of mechanisms to try.
1728 * dbus/dbus-auth.c (process_command): Remove check for lines
1729 longer that 1 MB; we only buffer up maximum 16 kB.
1731 * dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
1732 dbus/dbus-auth-script.c, dbus/dbus-auth.c, dbus/dbus-auth.h:
1733 Remove auth state AUTHENTICATED_WITH_UNUSED_BYTES, instead always
1734 assume there might be unused bytes.
1736 * dbus/dbus-auth.c (_dbus_auth_do_work): Remove check for
1737 client-out-of-mechs, it is handled in process_reject(). Move check
1738 for max failures to send_rejected(), as it's a server-only thing.
1740 * dbus/dbus-auth.c: Factor out protocol reply code into functions
1741 send_auth(), send_data(), send_rejected(), send_error(),
1742 send_ok(), send_begin() and send_cancel().
1744 2004-05-17 Kristian Høgsberg <krh@redhat.com>
1746 Remove base64 encoding, replace with hex encoding. Original patch
1747 from trow@ximian.com, added error handling.
1749 * dbus/dbus-string.c (_dbus_string_base64_encode)
1750 (_dbus_string_base64_decode): Remove.
1751 (_dbus_string_hex_decode): Add end_return argument so we can
1752 distinguish between OOM and invalid hex encoding.
1753 (_dbus_string_test): Remove base64 tests and add test case for
1756 * dbus/dbus-keyring.c, dbus/dbus-auth-script.c, dbus/dbus-auth.c:
1757 Replace base64 with hex.
1759 * test/data/auth/invalid-hex-encoding.auth-script: New test case
1760 for invalid hex encoded data in auth protocol.
1762 2004-05-17 Olivier Andrieu <oliv__a@users.sourceforge.net>
1764 * dbus/dbus-connection.c (check_for_reply_unlocked): plug a memory
1767 2004-05-15 Owen Fraser-Green <owen@discobabe.net>
1769 * mono/dbus-sharp.dll.config.in: Added for GAC
1770 * mono/dbus-sharp.snk: Added for GAC
1771 * mono/Assembly.cs.in: Added for GAC
1772 * mono/Makefile.am: Changes for GAC installation
1773 * configure.in: Added refs for dbus-sharp.dll.config.in and
1774 Assembly.cs.in. More fixes for mono testing
1775 * mono/example/Makefile.am: Changed var to CSC
1776 * Makefile.am: Changed flag name to DBUS_USE_CSC
1778 2004-05-15 Owen Fraser-Green <owen@discobabe.net>
1780 * mono/Makefile.am: Added SUBDIRS for docs. Changed SUBDIRS order
1781 * mono/doc/*: Added documentation framework
1782 * configure.in: Added monodoc check
1783 * README: Added description of mono configure flags
1785 2004-05-11 John (J5) Palmieri <johnp@redhat.com>:
1787 * doc/dbus-specification.xml: Added a "Required" column to the
1788 header fields table and changed the "zero or more" verbage in
1789 the above paragraph to read "The header must contain the required
1790 named header fields and zero or more of the optional named header
1792 * test/data/invalid-messages/*.message: Added the required PATH
1793 named header field to the tests so that they don't fail on
1794 'Missing path field'
1796 2004-05-07 John (J5) Palmieri <johnp@redhat.com>
1798 * python/dbus-bindings.pyx.in: Stopped the bindings from trashing
1799 the stack by implicitly defining variable and parameter types and
1800 removing the hack of defining C pointers as python objects and later
1803 2004-05-02 Owen Fraser-Green <owen@discobabe.net>
1805 * mono/Makefile.am: Removed test-dbus-sharp.exe from all target
1807 2004-05-01 Owen Fraser-Green <owen@discobabe.net>
1809 * mono/DBusType/Dict.cs: Handle empty dicts
1810 * mono/DBusType/Array.cs: Handle empty arrays
1811 * mono/Arguments.cs: Handle empty arguments
1813 2004-04-30 Owen Fraser-Green <owen@discobabe.net>
1815 * dbus-sharp.pc.in: Modified to include include Libs and Requires
1818 2004-04-25 Kristian Høgsberg <krh@redhat.com>
1820 * test/data/valid-messages/standard-*.message: Update message
1821 test scripts to new header field names.
1823 2004-04-22 John (J5) Palmieri <johnp@redhat.com>
1825 * test/break-loader.c (randomly_do_n_things): tracked down buffer
1826 overflow to times_we_did_each_thing array which would chop off the
1827 first character of the failure_dir string. Increased the size of
1828 the array to 7 to reflect the number of random mutation functions
1831 2004-04-21 Kristian Høgsberg <krh@redhat.com>
1833 * dbus/dbus-server-unix.c (unix_finalize): Don't unref
1834 unix_server->watch here, it is unreffed in disconnect.
1835 (_dbus_server_new_for_tcp_socket): convert NULL host to
1836 "localhost" here so we don't append NULL to address.
1838 * dbus/dbus-server.c (_dbus_server_test): Add test case for
1839 various addresses, including tcp with no explicit host.
1841 2004-04-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
1843 * dbus/dbus-message.c (decode_header_data, decode_string_field):
1844 fix incorrect setting of .name_offset in the HeaderField (it was
1845 off by two bytes, positioned right after the name and typecode)
1847 * bus/bus.c (bus_context_new, bus_context_unref): test before
1848 calling dbus_server_free_data_slot and _dbus_user_database_unref
1849 in case of an error.
1851 * tools/Makefile.am: add $(DBUS_GLIB_TOOL_LIBS), xml libs needed
1854 2004-04-19 Kristian Høgsberg <krh@redhat.com>
1856 * dbus/dbus-transport-unix.c (unix_do_iteration): Rewrite to use
1857 _dbus_poll() instead of select().
1859 2004-04-15 Jon Trowbridge <trow@ximian.com>
1861 * bus/main.c (signal_handler): Reload the configuration files
1863 (main): Set up our SIGHUP handler.
1865 * bus/bus.c (struct BusContext): Store the config file, user and
1866 fork flag in the BusContext.
1867 (process_config_first_time_only): Added. Contains the code
1868 (previously in bus_context_new) for setting up the BusContext from
1869 the BusConfigParser that should only be run the first time the
1870 config files are read.
1871 (process_config_every_time): Added. Contains the code (previously
1872 in bus_context_new) for setting up the BusContext from the
1873 BusConfigParser that should be run every time the config files are
1875 (load_config): Added. Builds a BusConfigParser from the config
1876 files and passes the resulting structure off to
1877 process_config_first_time_only (assuming this is the first time)
1878 and process_config_every_time.
1879 (bus_context_new): All of the config-related code has been moved
1880 to process_config_first_time_only and process_config_every_time.
1881 Now this function just does the non-config-related initializations
1882 and calls load_config.
1883 (bus_context_reload_config): Added.
1885 2004-04-15 Olivier Andrieu <oliv__a@users.sourceforge.net>
1887 * bus/driver.c (bus_driver_handle_get_service_owner):
1888 implement a GetServiceOwner method.
1889 * doc/dbus-specification.xml: document it.
1890 * dbus/dbus-errors.h: add a 'ServiceHasNoOwner' error.
1892 * glib/dbus-gproxy.c (dbus_gproxy_new_for_service_owner):
1893 implement, using the bus GetServiceOwner method.
1895 * test/glib/test-dbus-glib.c:
1896 use dbus_gproxy_new_for_service_owner so that we can receive the
1899 2004-04-15 John (J5) Palmieri <johnp@redhat.com>
1901 * dbus/dbus-internals.c, dbus/dbus-message-builder.c,
1902 dbus/dbus-message.c, dbus/dbus-protocol.h
1903 (DBUS_HEADER_FIELD_SERVICE): renamed DBUS_HEADER_FIELD_DESTINATION
1905 * dbus/dbus-internals.c, dbus/dbus-message-builder.c,
1906 dbus/dbus-message.c, dbus/dbus-protocol.h
1907 (DBUS_HEADER_FIELD_SENDER_SERVICE): renamed DBUS_HEADER_FIELD_SENDER
1909 * dbus/dbus-internals.c (_dbus_header_field_to_string):
1910 DBUS_HEADER_FIELD_DESTINATION resolves to "destination"
1911 DBUS_HEADER_FIELD_SENDER resolves to "sender"
1913 * doc/dbus-specification.xml (Header Fields Table):
1914 s/SERVICE/DESTINATION
1915 s/SENDER_SERVICE/SENDER
1918 2004-04-14 Olivier Andrieu <oliv__a@users.sourceforge.net>
1920 * test/glib/test-dbus-glib.c (timed_exit): fail the test after
1923 2004-04-13 Michael Meeks <michael@ximian.com>
1925 * glib/dbus-gobject.c (handle_introspect): split out
1926 (introspect_properties): this.
1927 (handle_introspect): implement this.
1929 * test/glib/Makefile.am: use the absolute path so the bus
1930 daemon's chdir ("/") doesn't kill us dead.
1932 * configure.in: subst ABSOLUTE_TOP_BUILDDIR
1934 2004-04-12 Jon Trowbridge <trow@ximian.com>
1936 * bus/config-parser.c (struct BusConfigParser): Added
1937 included_files field.
1938 (seen_include): Added. Checks whether or not a file has already
1939 been included by any parent BusConfigParser.
1940 (bus_config_parser_new): Copy the parent's included_files.
1941 (include_file): Track which files have been included, and fail on
1942 circular inclusions.
1943 (process_test_valid_subdir): Changed printf to report if we are
1944 testing valid or invalid conf files.
1945 (all_are_equiv): Changed printf to be a bit clearer about
1946 what we are actually doing.
1947 (bus_config_parser_test): Test invalid configuration files.
1949 2004-04-09 Jon Trowbridge <trow@ximian.com>
1951 * bus/config-parser.c (bus_config_parser_new): Added a 'parent'
1952 argument. If non-null, the newly-constructed BusConfigParser will
1953 be initialized with the parent's BusLimits instead of the default
1955 (include_file): When including a config file, pass in
1956 the current parser as the parent and then copy the BusLimits
1957 from the included BusConfigParser pack to the current parser.
1958 (process_test_valid_subdir): Renamed from process_test_subdir.
1959 (process_test_equiv_subdir): Added. Walks through a directory,
1960 descending into each subdirectory and loading the config files
1961 it finds there. If any subdirectory contains two config files
1962 that don't produce identical BusConfigParser structs, fail.
1963 For now, the BusConfigParser's BusPolicies are not compared.
1964 (bus_config_parser_test): Call both process_test_valid_subdir and
1965 process_test_equiv_subdir.
1967 * bus/config-loader-libxml.c (bus_config_load): Take a parent
1968 argument and pass it along to the call to bus_config_parser_new.
1969 Also made a few small changes to allow this code to compile.
1971 * bus/config-loader-expat.c (bus_config_load): Take a parent
1972 argument and pass it along to the call to bus_config_parser_new.
1974 * bus/bus.c (bus_context_new): Load the config file
1975 with a NULL parent argument.
1977 2004-03-29 Michael Meeks <michael@ximian.com>
1979 * glib/dbus-gobject.c (introspect_properties): split
1980 out, fix mangled 'while' flow control.
1981 (introspect_signals): implement.
1982 (handle_introspect): update.
1984 2004-03-29 Michael Meeks <michael@ximian.com>
1986 * glib/dbus-gobject.c (set_object_property): split out /
1987 re-work, use the property type, and not the message type(!)
1988 (get_object_property): ditto.
1990 * glib/dbus-gvalue.c (dbus_gvalue_demarshal),
1991 (dbus_gvalue_marshal): make this code re-usable, needed
1992 for signals too, also on both proxy and server side.
1993 Re-write for more efficiency / readability.
1995 2004-03-29 Michael Meeks <michael@ximian.com>
1997 * dbus/dbus-message.c
1998 (dbus_message_new_error_printf): impl.
2000 * dbus/dbus-connection.c
2001 (dbus_connection_unregister_object_path): fix warning.
2003 * configure.in: fix no-mono-installed situation.
2005 2004-03-27 Havoc Pennington <hp@redhat.com>
2007 Patch from Timo Teräs:
2009 * tools/dbus-send.c (main): if --print-reply, assume type is
2010 method call; support boolean type args
2012 * dbus/dbus-connection.c (dbus_connection_send_with_reply): fix a
2013 bunch of memleak and logic bugs
2015 2004-03-23 Owen Fraser-Green <owen@discobabe.net>
2017 * mono/Arguments.cs:
2018 * mono/Introspector.cs:
2020 * mono/InterfaceProxy.cs:
2022 * mono/ProxyBuilder.cs:
2024 Added InterfaceProxy class to avoid building proxies for every
2028 * dbus-message.c (dbus_message_append_args_valist)
2029 (dbus_message_iter_get_object_path)
2030 (dbus_message_iter_get_object_path_array)
2031 (dbus_message_iter_append_object_path)
2032 (dbus_message_iter_append_object_path_array):
2033 Added object_path iter functions to handle OBJECT_PATH arguments
2035 2004-03-23 Owen Fraser-Green <owen@discobabe.net>
2037 First checkin of mono bindings.
2040 Build stuff for the bindings
2041 * dbus-sharp.pc.in: Added for pkgconfig
2043 2004-03-21 Havoc Pennington <hp@redhat.com>
2045 * test/test-service.c (main): remove debug spew
2047 2004-03-21 Olivier Andrieu <oliv__a@users.sourceforge.net>
2049 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): accept empty
2052 * dbus/dbus-message.h, bus/dbus-message.c (dbus_message_iter_init)
2053 (dbus_message_iter_init_array_iterator)
2054 (dbus_message_iter_init_dict_iterator): return a dbus_bool_t to
2055 indicate whether the iterator is empty
2057 * dbus/dbus-pending-call.c, dbus/dbus-server.c: silence compiler
2060 2004-03-19 Havoc Pennington <hp@redhat.com>
2062 * NEWS: 0.21 updates
2064 * configure.in: 0.21
2066 * doc/Makefile.am: add all XMLTO usage to DBUS_XML_DOCS_ENABLED
2068 * python/Makefile.am: change to avoid dist of dbus_bindings.c so
2069 you don't need pyrex to make dist
2071 * qt/Makefile.am (libdbus_qt_1_la_SOURCES): add integrator.h to
2074 2004-03-18 Richard Hult <richard@imendio.com>
2076 * dbus/dbus-message.c (dbus_message_get_auto_activation)
2077 (dbus_message_set_auto_activation): Add doxygen docs.
2079 2004-03-16 Richard Hult <richard@imendio.com>
2081 * bus/activation.c: (bus_activation_service_created),
2082 (bus_activation_send_pending_auto_activation_messages),
2083 (bus_activation_activate_service):
2085 * bus/dispatch.c: (bus_dispatch),
2086 (check_nonexistent_service_auto_activation),
2087 (check_service_auto_activated),
2088 (check_segfault_service_auto_activation),
2089 (check_existent_service_auto_activation), (bus_dispatch_test):
2090 * bus/driver.c: (bus_driver_handle_activate_service):
2091 * bus/services.c: (bus_registry_acquire_service):
2092 * dbus/dbus-message.c: (dbus_message_set_auto_activation),
2093 (dbus_message_get_auto_activation):
2094 * dbus/dbus-message.h:
2095 * dbus/dbus-protocol.h: Implement auto-activation.
2097 * doc/dbus-specification.xml: Add auto-activation to the spec.
2099 2004-03-12 Olivier Andrieu <oliv__a@users.sourceforge.net>
2101 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos):
2102 fix a bug with CUSTOM types.
2104 * dbus/dbus-message.c (message_iter_test, _dbus_message_test): add
2105 a unit test for this bug (used to fail).
2107 2004-03-12 Mikael Hallendal <micke@imendio.com>
2110 (babysitter_watch_callback): notify all pending activations waiting for
2111 the same exec that the activation failed.
2112 (bus_activation_activate_service): shortcut the activation if we
2113 already waiting for the same executable to start up.
2115 2004-03-12 Mikael Hallendal <micke@imendio.com>
2118 - Added service file reloading.
2119 Each service files directory is kept in an hash table in
2120 BusActivation and each BusActivationEntry knows what .service-file it
2121 was read from. So when you try to activate a service the bus will
2122 check if it's been updated, removed or if new .service-files has
2124 - Test code at the bottom for the service file reloading.
2125 * bus/test-main.c: (main):
2127 - added service reloading test.
2128 * dbus/dbus-sysdeps.c:
2129 * dbus/dbus-sysdeps.h: (_dbus_delete_directory): Added.
2131 2004-03-08 Michael Meeks <michael@ximian.com>
2133 * dbus/dbus-connection.c (_dbus_connection_block_for_reply):
2134 bail immediately if disconnected, to avoid busy loop.
2136 * dbus/dbus-message.c (dbus_message_iter_get_args_valist):
2137 cleanup cut/paste/inefficiency.
2139 2004-03-01 David Zeuthen <david@fubar.dk>
2141 * dbus/dbus-string.c (_dbus_string_append_printf_valist): Fix a
2142 bug where args were used twice. This bug resulted in a segfault
2143 on a Debian/PPC system when starting the messagebus daemon. Include
2144 dbus-sysdeps.h for DBUS_VA_COPY
2146 * dbus/dbus-sysdeps.h: Define DBUS_VA_COPY if neccessary. From GLib
2148 * configure.in: Check for va_copy; define DBUS_VA_COPY to the
2149 appropriate va_copy implementation. From GLib
2151 2004-02-24 Joe Shaw <joe@ximian.com>
2153 * bus/services.c (bus_registry_acquire_service): We need to pass
2154 in the service name to dbus_set_error() to prevent a crash.
2156 2003-12-26 Anders Carlsson <andersca@gnome.org>
2158 * AUTHORS: Reveal my True identity.
2160 2003-12-17 Mikael Hallendal <micke@imendio.com>
2162 * dbus/dbus-message.c: (dbus_message_append_args_valist):
2163 - Added case for DBUS_TYPE_BYTE, patch from Johan Hedberg.
2165 2003-12-13 Mikael Hallendal <micke@imendio.com>
2167 * doc/TODO: Added not about better error check of configuration files.
2169 2003-12-02 Richard Hult <richard@imendio.com>
2171 * Update AFL version to 2.0 throughout the source files to reflect
2172 the update that was done a while ago.
2174 2003-12-02 Richard Hult <richard@imendio.com>
2176 * dbus/dbus-message.c (dbus_message_iter_append_dict): Set
2177 wrote_dict_key to FALSE on the iter that the dict is appended to,
2178 just like when appending other types. Fixes a bug where a dict
2179 couldn't be put inside a dict.
2180 (dbus_message_iter_append_dict_key): Fix typo in warning message.
2181 (message_iter_test, _dbus_message_test): Add test case for dict
2184 2003-12-01 David Zeuthen <david@fubar.dk>
2186 * python/dbus.py: Add the actual message when calling the reciever
2187 of a signal such that parameters can be inspected. Add the method
2188 remove_signal_receiver
2190 2003-11-26 Mikael Hallendal <micke@imendio.com>
2194 * glib/*.[ch]: Made ref functions return the pointer
2196 2003-11-25 Zack Rusin <zack@kde.org>
2198 * qt/integrator.h, qt/integrator.cpp: Adding handling of DBusServer,
2200 * qt/server.h, qt/server.cpp, qt/Makefile.am: Adding DBusServer
2203 * qt/connection.h, qt/connection.cpp: Adjusting to changes in
2204 the Integrator and to better fit with the server,
2206 2003-11-24 Zack Rusin <zack@kde.org>
2208 * qt/connection.h, qt/connection.cpp: removing initDbus method since
2209 the integrator handles it now
2211 * qt/integrator.h, qt/integrator.cpp: reworking handling of timeouts,
2212 since QTimer wasn't really meant to be used the way DBusTimeout is
2214 2003-11-24 Zack Rusin <zack@kde.org>
2216 * qt/integrator.h, qt/integrator.cpp, Makefile.am: Adding
2217 Integrator class which integrates D-BUS with the Qt event loop,
2219 * qt/connection.h, qt/connection.cpp: Move all the code which
2220 was dealing with D-BUS integration to the Integrator class,
2221 and start using Integrator,
2223 2003-11-23 Zack Rusin <zack@kde.org>
2225 * qt/connection.h, qt/connection.cpp: Adding the DBusConnection
2228 * qt/message.h, qt/message.cpp: updating to the current D-BUS api,
2229 switching namespaces to DBusQt, reworking the class,
2231 * Makefile.cvs: switching dependencies so that it matches KDE
2234 * qt/Makefile.am: adding connection.{h,cpp} and message.{h,cpp} to
2237 2003-11-19 Havoc Pennington <hp@redhat.com>
2241 * configure.in: bump version to 0.20
2243 * configure.in (have_qt): add yet another place to look for qt
2244 (someone hand trolltech a .pc file...)
2246 2003-11-01 Havoc Pennington <hp@redhat.com>
2248 * doc/dbus-specification.xml: add state machine docs on the auth
2249 protocol; just a first draft, I'm sure it's wrong.
2251 2003-10-28 David Zeuthen <david@fubar.dk>
2253 * python/dbus_bindings.pyx.in: add get_dict to handle dictionaries
2254 return types. Fixup TYPE_* to reflect changes in dbus/dbus-protocol.h
2256 2003-10-28 Havoc Pennington <hp@redhat.com>
2258 * dbus/dbus-message.c (get_next_field): delete unused function
2260 2003-10-28 Havoc Pennington <hp@redhat.com>
2262 * bus/expirelist.c (do_expiration_with_current_time): detect
2263 failure of the expire_func due to OOM
2265 * bus/connection.c (bus_pending_reply_expired): return FALSE on OOM
2267 * bus/dispatch.c (check_send_exit_to_service): fix to handle the
2268 NoReply error that's now created by the bus when the service exits
2270 2003-10-28 Havoc Pennington <hp@redhat.com>
2272 * dbus/dbus-message.c (_dbus_message_test): enable and fix the
2273 tests for set_path, set_interface, set_member, etc.
2275 * dbus/dbus-string.c (_dbus_string_insert_bytes): allow 0 bytes
2277 * dbus/dbus-message.c (set_string_field): always just delete and
2278 re-append the field; accept NULL for deletion
2279 (re_align_fields_recurse): reimplement
2281 2003-10-26 Havoc Pennington <hp@redhat.com>
2283 * dbus/dbus-connection.c: fix docs to properly describe the
2284 disconnected message
2285 (_dbus_connection_notify_disconnected): remove this function;
2286 we can't synchronously add the disconnected message, we have to
2287 do it after we've queued any remaining real messages
2288 (_dbus_connection_get_dispatch_status_unlocked): queue the
2289 disconnect message only if the transport has finished queueing all
2290 its real messages and is disconnected.
2291 (dbus_connection_disconnect): update the dispatch status here
2293 2003-10-22 Havoc Pennington <hp@redhat.com>
2295 * bus/bus.c (bus_context_check_security_policy): fix up assertion
2297 * bus/connection.c (bus_transaction_send_from_driver): set the
2298 destination to the connection's base service
2300 2003-10-20 Havoc Pennington <hp@redhat.com>
2302 hmm, make check is currently not passing.
2304 * doc/dbus-specification.xml: add requirement that custom type
2305 names follow the same rules as interface names.
2307 * dbus/dbus-protocol.h: change some of the byte codes, to avoid
2308 duplication and allow 'c' to be 'custom'; dict is now 'm' for
2311 * doc/dbus-specification.xml: update type codes to match
2312 dbus-protocol.h, using the ASCII byte values. Rename type NAMED to
2313 CUSTOM. Add type OBJECT_PATH to the spec.
2315 2003-10-17 Havoc Pennington <hp@redhat.com>
2317 * bus/driver.c (create_unique_client_name): use "." as separator
2318 in base service names instead of '-'
2320 * dbus/dbus-string.c (_dbus_string_get_byte): allow getting nul
2321 byte at the end of the string
2323 * dbus/dbus-internals.h (_DBUS_LIKELY, _DBUS_UNLIKELY): add
2324 optimization macros since string validation seems to be a slow
2327 * doc/dbus-specification.xml: restrict valid
2328 service/interface/member/error names. Add test suite code for the
2331 * dbus/dbus-string.c: limit service/interface/member/error names
2334 * dbus/dbus-connection.c (dbus_connection_dispatch): add missing
2335 format arg to verbose spew
2337 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): if not out of
2338 memory, return instead of g_error
2340 * test/test-service.c (path_message_func): support emitting a
2343 * dbus/dbus-bus.c (init_connections_unlocked): only fill in
2344 activation bus type if DBUS_BUS_ACTIVATION was set; default to
2345 assuming the activation bus was the session bus so that services
2346 started manually will still register.
2347 (init_connections_unlocked): fix so that in OOM situation we get
2348 the same semantics when retrying the function
2350 * test/test-service.c (main): change to use path registration, to
2351 test those codepaths; register with DBUS_BUS_ACTIVATION rather
2352 than DBUS_BUS_SESSION
2354 2003-10-16 Havoc Pennington <hp@redhat.com>
2356 * glib/dbus-gtest-main.c: bracket with #ifdef DBUS_BUILD_TESTS
2358 * Makefile.am (GCOV_DIRS): remove "test", we don't care about test
2359 coverage of the tests
2360 (coverage-report.txt): don't move the .da and .bbg files around
2362 2003-10-16 Havoc Pennington <hp@redhat.com>
2364 * bus/bus.c (struct BusContext): remove struct field I didn't mean
2367 2003-10-16 Havoc Pennington <hp@redhat.com>
2369 * bus/connection.c (bus_pending_reply_expired): either cancel or
2371 (bus_connections_check_reply): use unlink, not remove_link, as we
2372 don't want to free the link; fixes double free mess
2374 * dbus/dbus-pending-call.c (dbus_pending_call_block): fix in case
2375 where no reply was received
2377 * dbus/dbus-connection.c (_dbus_pending_call_complete_and_unlock):
2380 * bus/signals.c (match_rule_matches): add special cases for the
2381 bus driver, so you can match on sender/destination for it.
2383 * dbus/dbus-sysdeps.c (_dbus_abort): print backtrace if
2384 DBUS_PRINT_BACKTRACE is set
2386 * dbus/dbus-internals.c: add pid to assertion failure messages
2388 * dbus/dbus-connection.c: add message type code to the debug spew
2390 * glib/dbus-gproxy.c (gproxy_get_match_rule): match rules want
2391 sender=foo not service=foo
2393 * dbus/dbus-bus.c (dbus_bus_get): if the activation bus is the
2394 session bus but DBUS_SESSION_BUS_ADDRESS isn't set, use
2395 DBUS_ACTIVATION_ADDRESS instead
2397 * bus/activation.c: set DBUS_SESSION_BUS_ADDRESS,
2398 DBUS_SYSTEM_BUS_ADDRESS if appropriate
2400 * bus/bus.c (bus_context_new): handle OOM copying bus type into
2403 * dbus/dbus-message.c (dbus_message_iter_get_object_path): new function
2404 (dbus_message_iter_get_object_path_array): new function (half
2405 finished, disabled for the moment)
2407 * glib/dbus-gproxy.c (dbus_gproxy_end_call): properly handle
2408 DBUS_MESSAGE_TYPE_ERROR
2410 * tools/dbus-launch.c (babysit): support DBUS_DEBUG_OUTPUT to
2411 avoid redirecting stderr to /dev/null
2412 (babysit): close stdin if not doing the "exit_with_session" thing
2414 * dbus/dbus-sysdeps.c (_dbus_become_daemon): delete some leftover
2415 debug code; change DBUS_DEBUG_OUTPUT to only enable stderr, not
2416 stdout/stdin, so things don't get confused
2418 * bus/system.conf.in: fix to allow replies, I modified .conf
2419 instead of .conf.in again.
2421 2003-10-14 David Zeuthen <david@fubar.dk>
2423 * python/dbus_bindings.pyx.in (MessageIter.get): fixed typo in
2424 argtype to arg_type when raising unknown arg type exception.
2425 Changed type list to reflect the changes in dbus-protocol.h so
2426 the bindings actually work.
2428 2003-10-14 Havoc Pennington <hp@redhat.com>
2430 * test/decode-gcov.c: support gcc 3.3 also, though gcc 3.3 seems
2431 to have a bug keeping it from outputting the .da files sometimes
2432 (string_get_string): don't append garbage nul bytes to the string.
2434 2003-10-15 Seth Nickell <seth@gnome.org>
2436 * python/Makefile.am:
2438 Include dbus_h_wrapper.h in the dist tarball.
2440 2003-10-14 Havoc Pennington <hp@redhat.com>
2442 * bus/bus.c (bus_context_check_security_policy): revamp this to
2443 work more sanely with new policy-based requested reply setup
2445 * bus/connection.c (bus_transaction_send_from_driver): set bus
2446 driver messages as no reply
2448 * bus/policy.c (bus_client_policy_check_can_receive): handle a
2449 requested_reply attribute on allow/deny rules
2451 * bus/system.conf: add <allow requested_reply="true"/>
2453 * bus/driver.c (bus_driver_handle_message): fix check for replies
2454 sent to the bus driver, which was backward. How did this ever work
2455 at all though? I think I'm missing something.
2457 * dbus/dbus-message.c (decode_header_data): require error and
2458 method return messages to have a reply serial field to be valid
2459 (_dbus_message_loader_queue_messages): break up this function;
2460 validate that reply serial and plain serial are nonzero;
2461 clean up the OOM/error handling.
2462 (get_uint_field): don't return -1 from this
2463 (dbus_message_create_header): fix signed/unsigned bug
2465 * bus/connection.c (bus_connections_expect_reply): save serial of
2466 the incoming message, not reply serial
2468 2003-10-14 Havoc Pennington <hp@redhat.com>
2470 * bus/connection.c: implement pending reply tracking using
2473 * bus/bus.c (bus_context_check_security_policy): verify that a
2474 reply is pending in order to allow a reply to be sent. Deny
2475 messages of unknown type.
2477 * bus/dbus-daemon-1.1.in: update to mention new resource limits
2479 * bus/bus.c (bus_context_get_max_replies_per_connection): new
2480 (bus_context_get_reply_timeout): new
2482 2003-10-13 Seth Nickell <seth@gnome.org>
2484 * python/Makefile.am:
2486 Pass "make distcheck": remove a couple files from DIST_FILES
2487 that weren't included in the final version.
2489 2003-10-12 Havoc Pennington <hp@pobox.com>
2491 Added test code that 1) starts an actual bus daemon and 2) uses
2492 DBusGProxy; fixed bugs that were revealed by the test. Lots
2493 more testing possible, but this is the basic framework.
2495 * glib/dbus-gproxy.c (dbus_gproxy_manager_unregister): remove
2496 empty proxy lists from the proxy list hash
2498 * dbus/dbus-message.c (dbus_message_iter_get_args_valist): add a
2499 couple of return_if_fail checks
2501 * dbus/dbus-pending-call.c (_dbus_pending_call_new): use dbus_new0
2502 to allocate, so everything is cleared to NULL as it should be.
2504 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): pass
2505 source as data to dbus_connection_set_timeout_functions() as the
2506 timeout functions expected
2508 * test/glib/run-test.sh: add a little script to start up a message
2509 bus and run tests using it
2511 * tools/dbus-launch.1: updates
2513 * tools/dbus-launch.c (main): add --config-file option
2515 * tools/dbus-launch.c (main): remove confusing else if (runprog)
2516 that could never be reached.
2518 * dbus/dbus-message.c (dbus_message_new_method_return)
2519 (dbus_message_new_error, dbus_message_new_signal): set the
2520 no-reply-expected flag on all these. Redundant, but may
2521 as well be consistent.
2523 2003-10-11 Havoc Pennington <hp@pobox.com>
2525 * test/decode-gcov.c (function_solve_graph): make broken block
2526 graph a nonfatal error since it seems to be broken. Need to debug
2529 * dbus/dbus-marshal.c (_dbus_type_is_valid): new function since we
2530 can't just check type > INVALID < LAST anymore
2532 * dbus/dbus-message.c (dbus_message_get_signature): new function
2533 (dbus_message_has_signature): new function
2534 (struct DBusMessage): add signature field (right now it isn't sent
2535 over the wire, just generated on the fly)
2536 (dbus_message_copy): copy the signature, and init strings to
2537 proper length to avoid some reallocs
2538 (dbus_message_iter_init_array_iterator): return void, since it
2540 (dbus_message_iter_init_dict_iterator): return void since it can't fail
2541 (_dbus_message_loader_queue_messages): add silly temporary hack to
2542 fill in message->signature on load
2544 * dbus/dbus-protocol.h: change DBUS_TYPE_* values to be ASCII
2545 characters, so they are relatively human-readable.
2547 2003-10-11 Havoc Pennington <hp@pobox.com>
2549 * dbus/dbus-message.c (_dbus_message_test): add more test
2550 coverage, but #if 0 for now since they uncover a bug
2551 not fixed yet; I think in re_align_field_recurse()
2552 (re_align_field_recurse): add FIXME about broken assertion
2554 * dbus/dbus-sysdeps.c (_dbus_sysdeps_test): add more test coverage
2556 * bus/connection.c: share a couple code bits with expirelist.c
2558 * bus/expirelist.h, bus/expirelist.c: implement a generic
2559 expire-items-after-N-seconds facility, was going to share between
2560 expiring connections and replies, decided not to use for expiring
2561 connections for now.
2563 * COPYING: include AFL 2.0 (still need to change all the file headers)
2565 2003-10-09 Havoc Pennington <hp@redhat.com>
2567 * configure.in: define DBUS_HAVE_GCC33_GCOV if we have
2568 gcc 3.3. Not that we do anything about it yet.
2570 * bus/signals.c (bus_match_rule_parse): impose max length on the
2573 * dbus/dbus-protocol.h: add DBUS_MAXIMUM_MATCH_RULE_LENGTH
2575 2003-10-09 Havoc Pennington <hp@redhat.com>
2577 Make matching rules theoretically work (add parser).
2579 * bus/bus.c (bus_context_check_security_policy): fix up to handle
2580 the case where destination is explicitly specified as bus driver
2581 and someone else is eavesdropping.
2583 * bus/policy.c (bus_client_policy_check_can_receive): fix up
2584 definition of eavesdropping and assertion
2586 * tools/dbus-send.c (main): use dbus_message_type_from_string
2588 * bus/signals.c (bus_match_rule_parse): implement
2590 * dbus/dbus-message.c (dbus_message_type_from_string): new
2592 * dbus/dbus-errors.h (DBUS_ERROR_MATCH_RULE_INVALID): add
2594 2003-10-02 Havoc Pennington <hp@pobox.com>
2596 * glib/dbus-gproxy.c (dbus_gproxy_call_no_reply): rename from
2597 dbus_gproxy_oneway_call
2599 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main)
2600 (dbus_server_setup_with_g_main): fix to allow calling them more
2601 than once on the same args
2602 (dbus_bus_get_with_g_main): new function
2604 2003-10-02 Havoc Pennington <hp@redhat.com>
2606 * doc/dbus-tutorial.xml: write some stuff
2608 2003-09-29 Havoc Pennington <hp@pobox.com>
2610 * configure.in: split checks for Doxygen from XML docs, check for
2613 * doc/Makefile.am: XML-ify all the docs, and add a blank
2616 2003-09-29 Havoc Pennington <hp@pobox.com>
2618 * Merge dbus-object-names branch. To see the entire patch
2619 do cvs diff -r DBUS_OBJECT_NAMES_BRANCHPOINT -r dbus-object-names,
2620 it's huuuuge though.
2621 To revert, I tagged DBUS_BEFORE_OBJECT_NAMES_MERGE.
2623 2003-09-28 Havoc Pennington <hp@pobox.com>
2625 * HACKING: update to reflect new server
2627 2003-09-26 Seth Nickell <seth@gnome.org>
2630 * python/examples/example-signals.py:
2632 Start implementing some notions of signals. The API
2633 is really terrible, but they sort of work (with the
2634 exception of being able to filter by service, and to
2635 transmit signals *as* a particular service). Need to
2636 figure out how to make messages come from the service
2639 * python/dbus_bindings.pyx.in:
2641 Removed duplicate message_handler callbacks.
2643 2003-09-25 Havoc Pennington <hp@redhat.com>
2645 * bus/session.conf.in: fix my mess
2647 2003-09-25 Havoc Pennington <hp@pobox.com>
2649 * bus/session.conf.in: fix security policy, reported by Seth Nickell
2651 2003-09-25 Seth Nickell <seth@gnome.org>
2653 * python/examples/example-service.py:
2655 Johan notices complete wrong code in example-service, but
2656 completely wrong in a way that works exactly the same (!).
2657 Johan is confused, how could this possibly work? Example
2658 code fails to serve purpose of making things clear.
2661 2003-09-25 Mark McLoughlin <mark@skynet.ie>
2663 * doc/dbus-specification.sgml: don't require header fields
2664 to be 4-byte aligned and specify that fields should be
2665 distinguished from padding by the fact that zero is not
2668 * doc/TODO: remove re-alignment item and add item to doc
2669 the OBJECT_PATH type.
2671 * dbus/dbus-message.c:
2672 (HeaderField): rename the original member to value_offset
2673 and introduce a name_offset member to keep track of where
2674 the field actually begins.
2675 (adjust_field_offsets): remove.
2676 (append_int_field), (append_uint_field),
2677 (append_string_field): don't align the start of the header
2678 field to a 4-byte boundary.
2679 (get_next_field): impl finding the next marhsalled field
2680 after a given field.
2681 (re_align_field_recurse): impl re-aligning a number of
2682 already marshalled fields.
2683 (delete_field): impl deleting a field of any type and
2684 re-aligning any following fields.
2685 (delete_int_or_uint_field), (delete_string_field): remove.
2686 (set_int_field), (set_uint_field): no need to re-check
2687 that we have the correct type for the field.
2688 (set_string_field): ditto and impl re-aligning any
2690 (decode_header_data): update to take into account that
2691 the fields aren't 4-byte aligned any more and the new
2692 way to distinguish padding from header fields. Also,
2693 don't exit when there is too much header padding.
2694 (process_test_subdir): print the directory.
2695 (_dbus_message_test): add test to make sure a following
2696 field is re-aligned correctly after field deletion.
2698 * dbus/dbus-string.[ch]:
2699 (_dbus_string_insert_bytes): rename from insert_byte and
2700 allow the insert of multiple bytes.
2701 (_dbus_string_test): test inserting multiple bytes.
2703 * dbus/dbus-marshal.c: (_dbus_marshal_set_string): add
2704 warning note to docs about having to re-align any
2705 marshalled values following the string.
2707 * dbus/dbus-message-builder.c:
2708 (append_string_field), (_dbus_message_data_load):
2709 don't align the header field.
2711 * dbus/dbus-auth.c: (process_test_subdir): print the
2714 * test/break-loader.c: (randomly_add_one_byte): upd. for
2717 * test/data/invalid-messages/bad-header-field-alignment.message:
2720 * test/data/valid-messages/unknown-header-field.message: shove
2721 a dict in the unknown field.
2723 2003-09-25 Seth Nickell <seth@gnome.org>
2726 * python/dbus_bindings.pyx.in:
2728 Handle return values.
2730 * python/examples/example-client.py:
2731 * python/examples/example-service.py:
2733 Pass back return values from the service to the client.
2735 2003-09-24 Seth Nickell <seth@gnome.org>
2739 Connect Object methods (when you are sharing an object) up... pass
2740 in a list of methods to be shared. Sharing all the methods just
2741 worked out too weird. You can now create nice Services over the
2744 * python/dbus_bindings.pyx.in:
2746 Keep references to user_data tuples passed into C functions so
2747 Python doesn't garbage collect on us.
2749 Implement MethodReturn and Error subclasses of Message for creating
2750 DBusMessage's of those types.
2752 * python/examples/example-client.py:
2753 * python/examples/example-service.py:
2755 Simple example code showing both how create DBus services and objects,
2756 and how to use them.
2758 2003-09-23 Havoc Pennington <hp@pobox.com>
2760 * glib/dbus-gproxy.c (dbus_gproxy_manager_filter): implement
2762 2003-09-23 Havoc Pennington <hp@redhat.com>
2764 * glib/dbus-gproxy.c (dbus_gproxy_connect_signal): implement
2765 (dbus_gproxy_disconnect_signal): implement
2766 (dbus_gproxy_manager_remove_signal_match): implement
2767 (dbus_gproxy_manager_add_signal_match): implement
2768 (dbus_gproxy_oneway_call): implement
2770 2003-09-23 Havoc Pennington <hp@pobox.com>
2772 * glib/dbus-gproxy.c (struct DBusGProxy): convert to a GObject
2773 subclass. This means dropping the transparent thread safety of the
2774 proxy; you now need a separate proxy per-thread, or your own
2775 locking on the proxy. Probably right anyway.
2776 (dbus_gproxy_ref, dbus_gproxy_unref): nuke, just use g_object_ref
2778 2003-09-22 Havoc Pennington <hp@redhat.com>
2780 * glib/dbus-gproxy.c (dbus_gproxy_manager_get): implement
2782 2003-09-21 Seth Nickell <seth@gnome.org>
2784 First checkin of the Python bindings.
2786 * python/.cvsignore:
2787 * python/Makefile.am:
2788 * python/dbus_bindings.pyx.in:
2789 * python/dbus_h_wrapper.h:
2791 Pieces for Pyrex to operate on, building a dbus_bindings.so
2792 python module for low-level access to the DBus APIs.
2796 High-level Python module for accessing DBus objects.
2801 Build stuff for the python bindings.
2805 Extra macro needed for finding the Python C header files.
2807 2003-09-21 Havoc Pennington <hp@pobox.com>
2809 * glib/dbus-gproxy.c (dbus_gproxy_manager_new): start
2810 implementing the proxy manager, didn't get very far.
2812 * dbus/dbus-bus.c (dbus_bus_add_match): new
2813 (dbus_bus_remove_match): new
2815 * glib/dbus-gproxy.c (dbus_gproxy_new_for_service): add a
2816 path_name argument; adjust the other not-yet-implemented
2817 gproxy constructors to be what I think they should be.
2819 2003-09-21 Havoc Pennington <hp@pobox.com>
2821 * dbus/dbus-bus.c (dbus_bus_get): set exit_on_disconnect to TRUE
2822 by default for message bus connections.
2824 * dbus/dbus-connection.c (dbus_connection_dispatch): exit if
2825 exit_on_disconnect flag is set and we process the disconnected
2827 (dbus_connection_set_exit_on_disconnect): new function
2829 2003-09-21 Havoc Pennington <hp@pobox.com>
2831 Get matching rules mostly working in the bus; only actually
2832 parsing the rule text remains. However, the client side of
2833 "signal connections" hasn't been started, this patch is only the
2836 * dbus/dispatch.c: fix for the matching rules changes
2838 * bus/driver.c (bus_driver_handle_remove_match)
2839 (bus_driver_handle_add_match): send an ack reply from these
2842 * glib/dbus-gproxy.c (dbus_gproxy_begin_call): fix order of
2843 arguments, reported by Seth Nickell
2845 * bus/config-parser.c (append_rule_from_element): support
2846 eavesdrop=true|false attribute on policies so match rules
2847 can be prevented from snooping on the system bus.
2849 * bus/dbus-daemon-1.1.in: consistently use terminology "sender"
2850 and "destination" in attribute names; fix some docs bugs;
2851 add eavesdrop=true|false attribute
2853 * bus/driver.c (bus_driver_handle_add_match)
2854 (bus_driver_handle_remove_match): handle AddMatch, RemoveMatch
2857 * dbus/dbus-protocol.h (DBUS_SERVICE_ORG_FREEDESKTOP_BROADCAST): get
2858 rid of broadcast service concept, signals are just always broadcast
2860 * bus/signals.c, bus/dispatch.c, bus/connection.c, bus/bus.c:
2861 mostly implement matching rules stuff (currently only exposed as signal
2864 2003-09-21 Mark McLoughlin <mark@skynet.ie>
2866 * doc/dbus-specification.sgml: Change the header field name
2867 to be an enum and update the rest of the spec to reference
2868 the fields using the conventinal name.
2870 * dbus/dbus-protocol.h: update to reflect the spec.
2872 * doc/TODO: add item to remove the 4 byte alignment requirement.
2874 * dbus/dbus-message.c: Remove the code to generalise the
2875 header/body length and serial number header fields as named
2876 header fields so we can reference field names using the
2878 (append_int_field), (append_uint_field), (append_string_field):
2879 Append the field name as a byte rather than four chars.
2880 (delete_int_or_uint_field), (delete_string_field): reflect the
2881 fact that the field name and typecode now occupy 4 bytes instead
2883 (decode_string_field), (decode_header_data): update to reflect
2884 protocol changes and move the field specific encoding from
2885 decode_string_field() back into decode_header_data().
2887 * dbus/dbus-internals.[ch]: (_dbus_header_field_to_string):
2888 Add utility to aid debugging.
2890 * dbus/dbus-message-builder.c:
2891 (append_string_field), (_dbus_message_data_load): Update to
2892 reflect protocol changes; Change the FIELD_NAME directive
2893 to HEADER_FIELD and allow it to take the field's conventional
2894 name rather than the actual value.
2896 * test/data/*/*.message: Update to use HEADER_FIELD instead
2897 of FIELD_NAME; Always align the header on an 8 byte boundary
2898 *before* updating the header length.
2900 2003-09-15 Havoc Pennington <hp@pobox.com>
2902 * dbus/dbus-pending-call.c: add the get/set object data
2903 boilerplate as for DBusConnection, etc. Use generic object data
2904 for the notify callback.
2906 * glib/dbus-gparser.c (parse_node): parse child nodes
2908 * tools/dbus-viewer.c: more hacking on the dbus-viewer
2910 * glib/dbus-gutils.c (_dbus_gutils_split_path): add a file to
2911 contain functions shared between the convenience lib and the
2914 * glib/Makefile.am (libdbus_glib_1_la_LDFLAGS): add
2915 -export-symbols-regex to the GLib library
2917 * dbus/dbus-object-tree.c (_dbus_object_tree_dispatch_and_unlock):
2918 fix the locking in here, and add a default handler for
2919 Introspect() that just returns sub-nodes.
2921 2003-09-14 Havoc Pennington <hp@pobox.com>
2923 * glib/dbus-gthread.c (dbus_g_thread_init): rename to make g_foo
2924 rather than gfoo consistent
2926 * glib/dbus-gproxy.h: delete for now, move contents to
2927 dbus-glib.h, because the include files don't work right since we
2928 aren't in the dbus/ subdir.
2930 * glib/dbus-gproxy.c (dbus_gproxy_send): finish implementing
2931 (dbus_gproxy_end_call): finish
2932 (dbus_gproxy_begin_call): finish
2934 * glib/dbus-gmain.c (dbus_set_g_error): new
2936 * glib/dbus-gobject.c (handle_introspect): include information
2937 about child nodes in the introspection
2939 * dbus/dbus-connection.c (dbus_connection_list_registered): new
2940 function to help in implementation of introspection
2942 * dbus/dbus-object-tree.c
2943 (_dbus_object_tree_list_registered_and_unlock): new function
2945 2003-09-12 Havoc Pennington <hp@pobox.com>
2947 * glib/dbus-gidl.h: add common base class for all the foo_info
2950 * tools/dbus-viewer.c: add GTK-based introspection UI thingy
2953 * test/Makefile.am: try test srcdir -ef . in addition to test
2954 srcdir = ., one of them should work (yeah lame)
2956 * glib/Makefile.am: build the "idl" parser stuff as a convenience
2959 * glib/dbus-gparser.h: make description_load routines return
2960 NodeInfo* not Parser*
2962 * Makefile.am (SUBDIRS): build test dir after all library dirs
2964 * configure.in: add GTK+ detection
2966 2003-09-07 Havoc Pennington <hp@pobox.com>
2968 * Make Doxygen contented.
2970 2003-09-07 Havoc Pennington <hp@pobox.com>
2972 * doc/dbus-specification.sgml: more updates
2974 2003-09-06 Havoc Pennington <hp@pobox.com>
2976 * doc/dbus-specification.sgml: partial updates
2978 * bus/dbus-daemon-1.1.in: fix the config file docs for the
2979 zillionth time; hopefully I edited the right file this time.
2981 * bus/config-parser.c (append_rule_from_element): support
2982 send_type, send_path, receive_type, receive_path
2984 * bus/policy.c: add message type and path to the list of things
2985 that can be "firewalled"
2987 2003-09-06 Havoc Pennington <hp@pobox.com>
2989 * dbus/dbus-connection.c (dbus_connection_register_fallback): add this
2990 (dbus_connection_register_object_path): make this not handle
2991 messages to paths below the given path
2993 2003-09-03 Havoc Pennington <hp@pobox.com>
2995 * test/glib/Makefile.am: add this with random glib-linked test
2998 * glib/Makefile.am: remove the random test programs from here,
2999 leave only the unit tests
3001 * glib/dbus-gobject.c (_dbus_gobject_test): add test for
3002 uscore/javacaps conversion, and fix
3003 (get_object_property, set_object_property): change to .NET
3004 convention for mapping props to methods, set_FooBar/get_FooBar,
3005 since one language has such a convention we may as well copy it.
3006 Plus real methods in either getFooBar or get_foo_bar style won't
3007 collide with this convention.
3009 2003-09-01 Havoc Pennington <hp@pobox.com>
3011 * glib/dbus-gparser.c: implement
3013 * glib/dbus-gobject.c: start implementing skeletons support
3015 * configure.in: when disabling checks/assert, also define
3016 G_DISABLE_ASSERT and G_DISABLE_CHECKS
3018 2003-09-01 Havoc Pennington <hp@pobox.com>
3020 * glib/Makefile.am: rearrange a bunch of files and get "make
3021 check" framework set up
3023 2003-08-31 Havoc Pennington <hp@pobox.com>
3025 * fix build with --disable-tests
3027 2003-08-30 Havoc Pennington <hp@pobox.com>
3029 * dbus/dbus-connection.c: purge DBusMessageHandler
3031 * dbus/dbus-message-handler.c: remove DBusMessageHandler, just
3032 use callbacks everywhere
3034 2003-08-30 Havoc Pennington <hp@pobox.com>
3036 * test/data/valid-config-files/system.d/test.conf: change to
3037 root for the user so warnings don't get printed
3039 * dbus/dbus-message.c: add dbus_message_get_path,
3040 dbus_message_set_path
3042 * dbus/dbus-object-tree.c (do_test_dispatch): add test of
3043 dispatching to a path
3045 * dbus/dbus-string.c (_dbus_string_validate_path): add
3047 * dbus/dbus-marshal.c (_dbus_demarshal_object_path): implement
3048 (_dbus_marshal_object_path): implement
3050 * dbus/dbus-protocol.h (DBUS_HEADER_FIELD_PATH): new header field
3051 to contain the path to the target object
3052 (DBUS_HEADER_FIELD_SENDER_SERVICE): rename
3053 DBUS_HEADER_FIELD_SENDER to explicitly say it's the sender service
3055 2003-08-30 Havoc Pennington <hp@pobox.com>
3057 * dbus/dbus-object-tree.c: write tests and fix the discovered bugs
3059 2003-08-29 Havoc Pennington <hp@pobox.com>
3061 * dbus/dbus-object-tree.c: modify to allow overlapping paths to be
3063 (struct DBusObjectSubtree): shrink this
3064 a lot, since we may have a lot of them
3065 (_dbus_object_tree_free_all_unlocked): implement
3066 (_dbus_object_tree_dispatch_and_unlock): implement
3068 2003-08-29 Havoc Pennington <hp@pobox.com>
3070 * dbus/dbus-internals.h: fix _DBUS_N_GLOBAL_LOCKS
3072 2003-08-28 Havoc Pennington <hp@pobox.com>
3076 * dbus/dbus-connection.c: port to no ObjectID, create a
3077 DBusObjectTree, rename ObjectTree to ObjectPath in public API
3079 * dbus/dbus-connection.h (struct DBusObjectTreeVTable): delete
3080 everything except UnregisterFunction and MessageFunction
3082 * dbus/dbus-marshal.c: port away from DBusObjectID,
3083 add DBUS_TYPE_OBJECT_PATH
3085 * dbus/dbus-object-registry.[hc], dbus/dbus-object.[hc],
3086 dbus/dbus-objectid.[hc]: remove these, we are moving to
3087 path-based object IDs
3089 2003-08-25 Havoc Pennington <hp@pobox.com>
3091 Just noticed that dbus_message_test is hosed, I wonder when I
3092 broke that. I thought make check was passing earlier...
3094 * dbus/dbus-object-tree.c: add new "object tree" to match DCOP
3095 container tree, will replace most of dbus-object-registry
3097 * dbus/dbus-string.c (_dbus_string_append_printf_valist): fix C99
3100 2003-08-19 Havoc Pennington <hp@pobox.com>
3102 * dbus/dbus-message.c (decode_string_field): support FIELD_SENDER
3103 (dbus_message_is_error): fix this function
3105 * bus/dbus-daemon-1.1: clarify logic on when <deny>/<allow> rules
3108 * bus/policy.c (bus_client_policy_check_can_receive): fix code to
3109 reflect clarified man page
3110 (bus_client_policy_check_can_send): ditto
3112 * bus/session.conf.in: fixup
3114 * bus/system.conf.in: fixup
3116 2003-08-18 Havoc Pennington <hp@redhat.com>
3118 * dbus/dbus-hash.c (_dbus_hash_table_insert_two_strings): fix
3120 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
3121 dumb bug created earlier (wrong order of args to
3122 decode_header_data())
3124 * tools/dbus-send.c: port
3126 * tools/dbus-print-message.c (print_message): port
3128 * test/data/*messages: port all messages over
3130 * dbus/dbus-message-builder.c: support including
3133 * bus/driver.c: port over
3135 * bus/dispatch.c: port over to new stuff
3137 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
3138 rename disconnect signal to "Disconnected"
3140 2003-08-17 Havoc Pennington <hp@pobox.com>
3142 This doesn't compile yet, but syncing up so I can hack on it from
3143 work. What are branches for if not broken code? ;-)
3145 * dbus/dbus-protocol.h: remove DBUS_HEADER_FIELD_NAME, add
3146 DBUS_HEADER_FIELD_INTERFACE, DBUS_HEADER_FIELD_MEMBER,
3147 DBUS_HEADER_FIELD_ERROR_NAME
3149 * dbus/dbus-hash.c: Introduce DBUS_HASH_TWO_STRINGS as hack to use
3150 for the interface+member pairs
3151 (string_hash): change to use g_str_hash algorithm
3152 (find_direct_function, find_string_function): refactor these to
3155 * dbus/dbus-message.c: port all of this over to support
3156 interface/member fields instead of name field
3158 * dbus/dbus-object-registry.c: port over
3160 * dbus/dbus-string.c (_dbus_string_validate_interface): rename
3161 from _dbus_string_validate_name
3163 * bus/dbus-daemon-1.1: change file format for the
3164 <deny>/<allow> stuff to match new message naming scheme
3166 * bus/policy.c: port over
3168 * bus/config-parser.c: parse new format
3170 2003-08-16 Havoc Pennington <hp@pobox.com>
3172 * dbus/dbus-object-registry.c (add_and_remove_objects): remove
3175 * glib/dbus-gproxy.c: some hacking
3177 2003-08-15 Havoc Pennington <hp@redhat.com>
3179 * dbus/dbus-pending-call.c (dbus_pending_call_block): implement
3181 * dbus/dbus-connection.c
3182 (dbus_connection_send_with_reply_and_block): factor out internals;
3183 change to convert any error replies to DBusError instead of
3184 returning them as a message
3186 2003-08-15 Havoc Pennington <hp@pobox.com>
3188 * dbus/dbus-connection.c,
3189 dbus/dbus-pending-call.c: Finish the pending call stuff
3191 2003-08-14 Havoc Pennington <hp@redhat.com>
3193 * dbus/dbus-pending-call.c: start on new object that will replace
3194 DBusMessageHandler and ReplyHandlerData for tracking outstanding
3197 * dbus/dbus-gproxy.c: start on proxy object used to communicate
3198 with remote interfaces
3200 * dbus/dbus-gidl.c: do the boring boilerplate in here
3202 2003-08-12 Havoc Pennington <hp@pobox.com>
3204 * bus/dispatch.c (bus_dispatch): make this return proper
3205 DBusHandlerResult to avoid DBUS_ERROR_UNKNOWN_METHOD
3207 * dbus/dbus-errors.c (dbus_set_error): use
3208 _dbus_string_append_printf_valist
3210 * dbus/dbus-string.c (_dbus_string_append_printf_valist)
3211 (_dbus_string_append_printf): new
3213 * dbus/dbus-errors.h (DBUS_ERROR_UNKNOWN_MESSAGE): change to
3216 * dbus/dbus-connection.c (dbus_connection_dispatch): handle
3217 DBUS_HANDLER_RESULT_NEED_MEMORY; send default error reply if a
3218 message is unhandled.
3220 2003-08-11 Havoc Pennington <hp@pobox.com>
3222 * bus/test.c (client_disconnect_handler): change to return
3223 HANDLED (would have been REMOVE_MESSAGE)
3225 * dbus/dbus-object.h (enum DBusHandlerResult): rename to
3226 HANDLED/NOT_YET_HANDLED instead of
3227 REMOVE_MESSAGE/ALLOW_MORE_HANDLERS to make it clearer how it
3230 2003-08-10 Havoc Pennington <hp@pobox.com>
3232 * tools/dbus-send.c (main): add --type argument, for now
3233 supporting only method_call and signal types.
3235 * tools/dbus-print-message.c: print message type
3237 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
3238 init connection->objects
3240 * doc/dbus-specification.sgml: fix sgml
3242 * bus/*.c: port over to object-instance API changes
3244 * test/test-service.c: ditto
3246 * dbus/dbus-message.c (dbus_message_create_header): allow #NULL
3247 name, we will have to fix up the rest of the code to also handle
3249 (dbus_message_new): generic message-creation call
3250 (set_string_field): allow appending name field
3252 2003-08-06 Havoc Pennington <hp@pobox.com>
3254 * dbus/dbus-object-registry.c: implement signal connection
3257 * dbus/dbus-connection.c (_dbus_connection_unref_unlocked): new
3259 * dbus/dbus-internals.c (_dbus_memdup): new function
3261 2003-08-02 Havoc Pennington <hp@pobox.com>
3263 * dbus/dbus-message.c (dbus_message_get_no_reply)
3264 (dbus_message_set_no_reply): add these and remove
3265 set_is_error/get_is_error
3267 * dbus/dbus-protocol.h, doc/dbus-specification.sgml:
3268 remove the ERROR flag, since there's now an ERROR type
3270 2003-08-01 Havoc Pennington <hp@pobox.com>
3272 * dbus/dbus-object-registry.c (_dbus_object_registry_handle_and_unlock):
3275 * dbus/dbus-message.c (dbus_message_get_type): new function
3277 * doc/dbus-specification.sgml: add "type" byte to messages
3279 2003-08-01 Havoc Pennington <hp@pobox.com>
3281 * dbus/dbus-protocol.h (DBUS_MESSAGE_TYPE_*): introduce
3282 a message type enum to distinguish kinds of message
3283 (DBUS_HEADER_FLAG_NO_REPLY_EXPECTED): flag for a message
3284 that need not be replied to
3286 2003-08-01 Havoc Pennington <hp@pobox.com>
3288 * dbus/dbus-marshal.c: adapt to DBusObjectID changes
3289 (unpack_8_octets): fix no-64-bit-int bug
3291 * dbus/dbus-object-registry.c (validate_id): validate the
3292 connection ID bits, not just the instance ID.
3294 * dbus/dbus-connection.c (_dbus_connection_init_id): initialize
3295 the connection-global 33 bits of the object ID
3297 * dbus/dbus-object-registry.c (info_from_entry): fill in
3298 object ID in the new way
3300 * dbus/dbus-objectid.h: rather than high/low bits, specifically
3301 define server/client/instance bits.
3303 2003-07-30 Havoc Pennington <hp@pobox.com>
3305 * dbus/dbus-connection.c (dbus_connection_register_object): fix
3308 2003-07-13 Havoc Pennington <hp@pobox.com>
3310 * dbus/dbus-object.h (struct DBusObjectVTable): add padding
3311 fields to DBusObjectVTable and DBusObjectInfo
3313 2003-07-12 Havoc Pennington <hp@pobox.com>
3315 * dbus/dbus-object-registry.c: implement unit test,
3316 fix bugs discovered in process
3318 * dbus/dbus-connection.c: remove handler_table and
3319 register_handler(), add DBusObjectRegistry usage
3321 * dbus/dbus-objectid.c (dbus_object_id_is_null)
3322 (dbus_object_id_set_null): new functions
3324 2003-07-08 Havoc Pennington <hp@pobox.com>
3326 * dbus/dbus-object.c: implement some of this
3328 * dbus/dbus-object-registry.c
3329 (_dbus_object_registry_add_and_unlock): fill in the object_id out
3331 (_dbus_object_registry_new): handle OOM
3333 2003-07-08 Havoc Pennington <hp@pobox.com>
3335 * dbus/dbus-object.h: sketch out an API for registering objects
3336 with a connection, that allows us to use as little as 24 bytes
3337 per object and lets application code represent an object in
3338 any conceivable way.
3340 * dbus/dbus-object-registry.c: implement the hard bits of the
3341 DBusConnection aspect of object API. Not yet wired up.
3343 2003-07-06 Havoc Pennington <hp@pobox.com>
3345 * dbus/dbus-marshal.c (_dbus_marshal_set_object_id): new function
3346 (_dbus_marshal_object_id): new
3347 (_dbus_demarshal_object_id): new
3348 (_dbus_marshal_get_arg_end_pos): support object ID type, and
3349 consolidate identical switch cases. Don't conditionalize handling
3350 of DBUS_TYPE_UINT64, need to handle the type always.
3351 (_dbus_marshal_validate_arg): consolidate identical cases, and
3352 handle DBUS_TYPE_OBJECT_ID
3354 * dbus/dbus-objectid.c: new file with DBusObjectID data type.
3356 * dbus/dbus-protocol.h: add DBUS_TYPE_OBJECT_ID
3358 2003-09-28 Havoc Pennington <hp@pobox.com>
3362 2003-09-28 Havoc Pennington <hp@pobox.com>
3364 * doc/Makefile.am (dbus-specification.html): testing a funky hack
3365 to work with Debian db2html
3367 2003-09-28 Havoc Pennington <hp@pobox.com>
3369 * configure.in: 0.13
3371 * doc/Makefile.am (dbus-test-plan.html): accept nonexistence of
3372 stylesheet-images for benefit of Debian
3374 Change back to using filesystem-linked sockets for the system
3375 bus, so only root can create the default system bus address.
3377 * bus/system.conf.in: change to use
3378 DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
3380 * dbus/Makefile.am (INCLUDES): remove DBUS_SYSTEM_BUS_PATH define
3383 * configure.in: define DBUS_SYSTEM_BUS_DEFAULT_ADDRESS
3384 here, and AC_DEFINE DBUS_SYSTEM_PATH
3386 2003-08-09 Anders Carlsson <andersca@codefactory.se>
3389 * doc/busconfig.dtd:
3392 2003-08-09 Anders Carlsson <andersca@codefactory.se>
3394 * doc/dbus-specification.sgml:
3395 Add activation reply values.
3397 2003-08-05 Havoc Pennington <hp@redhat.com>
3399 * configure.in: 0.12
3401 2003-08-05 Anders Carlsson <andersca@codefactory.se>
3403 * glib/dbus-gmain.c: (watch_fd_new), (watch_fd_ref),
3404 (watch_fd_unref), (dbus_gsource_check), (dbus_gsource_dispatch),
3405 (add_watch), (remove_watch), (create_source):
3406 Refcount fds, fixes some reentrancy issues.
3408 2003-07-30 Havoc Pennington <hp@redhat.com>
3410 * dbus/dbus-bus.c (init_connections_unlocked): fix default system
3411 bus address to be abstract if we have abstract sockets
3415 2003-07-28 Havoc Pennington <hp@redhat.com>
3417 * bus/messagebus.in: fix to avoid processname/servicename
3418 confusion, from Michael Kearey
3419 https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=100965
3421 2003-07-23 Havoc Pennington <hp@pobox.com>
3423 * dbus/dbus-message.c (dbus_message_iter_get_named):
3424 fix from Andy Hanton to remove broken "+1"
3426 2003-07-16 Havoc Pennington <hp@pobox.com>
3428 * tools/dbus-launch.c (babysit): close stdout/stderr in the
3429 babysitter process, as suggested by Thomas Leonard, so
3430 an "eval `dbus-launch --exit-with-session`" will actually
3433 2003-07-16 Havoc Pennington <hp@pobox.com>
3435 * configure.in: print out EXPANDED_* variables in the summary at
3436 the end; clean up the code that computes EXPANDED_ variables and
3437 get the ones using exec_prefix right. Should make things work
3438 when you build without --prefix
3440 2003-06-29 Havoc Pennington <hp@pobox.com>
3442 * mono/Test.cs (class Test): fire up a main loop and run it
3444 * mono/DBus.cs (DBus): don't g_thread_init since it can only be
3445 done once, the app has to do it
3447 2003-06-26 Havoc Pennington <hp@pobox.com>
3449 * mono/Connection.cs: set up connection with the glib main loop
3451 2003-07-01 Havoc Pennington <hp@redhat.com>
3453 * doc/dbus-specification.sgml: clarify the format of a type code,
3454 change suggested by Jim Blandy
3456 2003-06-29 Miloslav Trmac <mitr@volny.cz>
3459 * tools/Makefile.am: Don't assume srcdir == builddir.
3461 * dbus/dbus-memory.c (dbus_realloc): Don't check guards after shrinking
3462 the allocated block.
3463 (_dbus_memory_test): New function.
3464 * dbus/dbus-test.h: Add _dbus_memory_test ().
3465 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): Call it.
3467 * dbus/dbus-message.c (decode_header_data): Use %.4s instead
3469 (dbus_message_new): Remove obsolete @todo.
3471 * dbus/dbus-marshal.c (_dbus_marshal_set_int64)
3472 (_dbus_marshal_set_uint64): Fix comment.
3474 * dbus/dbus-message.c (append_int_field, append_uint_field): Don't
3475 hardcode FIELD_REPLY_SERIAL.
3477 * dbus/dbus-mainloop.c (_dbus_loop_remove_watch)
3478 (_dbus_loop_remove_timeout): Cast function pointers to (void *) for %p
3480 * configure.in: Add -D_POSIX_C_SOURCE=199309L -DBSD_SOURCE to CFLAGS
3481 and disable DBUS_USE_ATOMIC_INT_486 when --enable-ansi is used
3483 2003-06-24 Havoc Pennington <hp@pobox.com>
3485 * mono/*.cs: Use IntPtr.Zero instead of ((IntPtr) 0)
3487 2003-06-23 Anders Carlsson <andersca@codefactory.se>
3493 * gcj/TestMessage.java: (TestMessage), (TestMessage.main):
3494 * gcj/org/.cvsignore:
3495 * gcj/org/Makefile.am:
3496 * gcj/org/freedesktop/.cvsignore:
3497 * gcj/org/freedesktop/Makefile.am:
3498 * gcj/org/freedesktop/dbus/.cvsignore:
3499 * gcj/org/freedesktop/dbus/Makefile.am:
3500 * gcj/org/freedesktop/dbus/Message.java: (Message),
3502 * gcj/org/freedesktop/dbus/natMessage.cc:
3503 Fix the build system.
3505 2003-06-22 Havoc Pennington <hp@pobox.com>
3507 * mono/Connection.cs: add more bindings
3509 * dbus/dbus-threads.c (dbus_threads_init): allow calling this
3512 2003-06-22 Havoc Pennington <hp@pobox.com>
3514 * mono/Connection.cs, mono/DBus.cs, mono/Error.cs:
3515 Start wrapping more stuff.
3517 2003-06-22 Havoc Pennington <hp@pobox.com>
3519 * mono/Message.cs: implement Message.Wrap() that ensures we only
3520 have a single C# wrapper per DBusMessage, assuming it works which
3521 it probably doesn't.
3523 * dbus/dbus-message.c (dbus_message_allocate_data_slot): new
3524 (dbus_message_free_data_slot): new
3525 (dbus_message_set_data): new
3526 (dbus_message_get_data): new
3528 2003-06-22 Havoc Pennington <hp@pobox.com>
3530 * dbus/dbus-dataslot.c (_dbus_data_slot_allocator_unref)
3531 (_dbus_data_slot_allocator_alloc): rework these to keep a
3532 reference count on each slot and automatically manage a global
3533 slot ID variable passed in by address
3535 * bus/bus.c: convert to new dataslot API
3537 * dbus/dbus-bus.c: convert to new dataslot API
3539 * dbus/dbus-connection.c: convert to new dataslot API
3541 * dbus/dbus-server.c: convert to new dataslot API
3543 * glib/dbus-gmain.c: ditto
3547 * bus/connection.c: ditto
3549 2003-06-22 Anders Carlsson <andersca@codefactory.se>
3551 * configure.in: Add AM_PROG_GCJ and move AM_PROG_LIBTOOL
3552 after the gcj checks so that the correct configuration tags
3553 will be added to libtool.
3555 * dbus-glib-1.pc.in: No need to specify any includes since
3556 dbus-1.pc.in has those.
3558 2003-06-22 Havoc Pennington <hp@pobox.com>
3560 * mono/*, gcj/*, configure.in, Makefile.am:
3561 Check in makefiles and subdirs for mono and gcj bindings.
3562 Neither binding actually exists, just trying to get through
3563 all the build and other boring bits.
3565 2003-06-21 Philip Blundell <philb@gnu.org>
3567 * tools/dbus-monitor.1: Updated.
3569 * tools/dbus-send.1: Likewise.
3571 2003-06-20 Anders Carlsson <andersca@codefactory.se>
3573 * dbus/dbus-transport-unix.c (unix_handle_watch): Check
3574 for hangup and error after checking read so we won't discard
3575 pending data if both hangup and read are set.
3577 2003-06-19 Philip Blundell <philb@gnu.org>
3579 * tools/dbus-print-message.c (print_message): Handle BOOLEAN.
3581 * tools/dbus-send.c: Accept both --system and --session.
3583 * tools/dbus-monitor.c: Same here.
3585 2003-06-19 Anders Carlsson <andersca@codefactory.se>
3587 * glib/dbus-glib.h: Fix so that dbus-glib.h can be used
3588 from C++ (Patch by Miloslav Trmac).
3590 2003-06-15 Joe Shaw <joe@assbarn.com>
3592 * configure.in: Check for socklen_t.
3594 * dbus/dbus-sysdeps.c: Define socklen_t if it's not defined.
3596 * test/test-segfault.c: Add #include <sys/time.h>
3598 * tools/Makefile.am: Add DBUS_X_CFLAGS to the INCLUDES since
3599 dbus-launch needs it.
3601 2003-06-09 Havoc Pennington <hp@redhat.com>
3603 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): don't use
3604 SUN_LEN, it breaks abstract socket usage
3606 * dbus/dbus-internals.c (_dbus_verbose_real): only print PID at
3609 2003-06-04 Havoc Pennington <hp@pobox.com>
3611 * dbus/dbus-server.c (dbus_server_listen): allow abstract sockets
3612 using unix:abstract=/foo, and when listening in a tmpdir
3613 i.e. unix:tmpdir=/tmp, always use abstract sockets if we can.
3615 * dbus/dbus-transport.c (_dbus_transport_open): support
3618 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
3619 support abstract sockets
3621 * dbus/dbus-transport-unix.c
3622 (_dbus_transport_new_for_domain_socket): support abstract sockets
3624 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket): add "abstract"
3625 toggle as an argument, implement abstract namespace support
3626 (_dbus_listen_unix_socket): ditto
3628 * configure.in: add --enable-abstract-sockets and implement
3629 a configure check for autodetection of the right value.
3631 2003-06-01 Havoc Pennington <hp@pobox.com>
3633 * tools/dbus-cleanup-sockets.c: add utility to clean up sockets
3634 in /tmp (though on Linux this will end up being useless,
3635 when we add abstract namespace support)
3637 * configure.in: define DBUS_SESSION_SOCKET_DIR in addition to
3640 2003-05-28 Colin Walters <walters@verbum.org>
3642 * tools/dbus-monitor.c (main): Fix silly typo (s/--session/--system/).
3644 2003-05-18 Anders Carlsson <andersca@codefactory.se>
3646 * dbus/dbus-message.c (dbus_message_new): Remove @todo.
3648 2003-05-17 Colin Walters <walters@gnu.org>
3650 * tools/dbus-send.c: Don't exit with an error code if --help was
3651 passed. Default to using the session bus instead of the system
3654 * tools/dbus-launch.c: Ditto.
3656 * tools/dbus-monitor.c: Ditto.
3658 * tools/dbus-send.1: Update with new arguments.
3660 * tools/dbus-launch.c: Emit code to export variables. New
3661 arguments -s and -c to specify shell syntax, and a bit of code to
3662 autodetect syntax. Also, allow specifying a program to run.
3664 * tools/dbus-launch.1: Update with new arguments.
3666 * tools/dbus-send.1: Ditto.
3668 * tools/dbus-monitor.1: Ditto.
3670 2003-05-17 Havoc Pennington <hp@pobox.com>
3672 * bus/config-parser.c (merge_included): merge in policies from
3673 child configuration file.
3675 * bus/policy.c (bus_policy_merge): function to merge two policies
3678 2003-05-16 Havoc Pennington <hp@redhat.com>
3680 * dbus/dbus-connection.c: disable verbose lock spew
3682 * tools/dbus-send.c: add --print-reply command line option
3684 * tools/dbus-print-message.h (print_message): new util function
3685 shared by dbus-send and dbus-monitor
3687 * tools/dbus-monitor.c (handler_func): exit on disconnect
3689 * dbus/dbus-transport-unix.c (do_reading): if the transport is
3690 disconnected, don't try to use the read_watch
3692 * dbus/dbus-watch.c (dbus_watch_get_enabled): assert watch != NULL
3693 so we can find this bug more easily
3695 2003-05-16 Havoc Pennington <hp@redhat.com>
3697 * bus/policy.c (free_rule_list_func): avoid a crash when passed
3698 NULL as DBusHashTable is annoyingly likely to do.
3700 2003-05-16 Colin Walters <walters@verbum.org>
3702 * tools/dbus-monitor.c: Add --session argument and usage()
3705 * tools/dbus-monitor.1: Update with new --session arg.
3707 * bus/Makefile.am (install-data-hook): Create
3708 $(libdir)/dbus-1.0/services so that the session bus is happy.
3710 2003-05-15 Havoc Pennington <hp@redhat.com>
3712 * dbus/dbus-sysdeps.c (_dbus_atomic_dec, _dbus_atomic_inc): work
3713 on non-x86. ifdef's are evil.
3715 2003-05-15 Havoc Pennington <hp@redhat.com>
3717 * configure.in: 0.11
3721 * bus/Makefile.am (initddir): apparently we are supposed to put
3722 init scripts in /etc/rc.d/init.d not /etc/init.d
3724 * bus/Makefile.am: remove the "you must --enable-tests to make
3725 check" as it broke distcheck
3727 * bus/Makefile.am (install-data-hook): create /etc/dbus-1/system.d
3729 2003-05-13 James Willcox <jwillcox@gnome.org>
3732 * bus/activation.c: (bus_activation_service_created),
3733 (bus_activation_activate_service):
3734 * bus/driver.c: (bus_driver_send_service_deleted),
3735 (bus_driver_send_service_created), (bus_driver_send_service_lost),
3736 (bus_driver_send_service_acquired),
3737 (bus_driver_send_welcome_message),
3738 (bus_driver_handle_list_services):
3739 * bus/session.conf.in:
3740 * dbus/dbus-bus.c: (dbus_bus_acquire_service),
3741 (dbus_bus_service_exists), (dbus_bus_activate_service):
3744 Add some convenience API which lets you activate a service, and did a
3745 bunch of s/0/DBUS_TYPE_INVALID/ in calls to dbus_message_append_args()
3746 and dbus_message_get_args()
3748 2003-05-11 Havoc Pennington <hp@pobox.com>
3750 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix to avoid
3751 calling _dbus_marshal_validate_arg() for every byte in a byte
3754 * dbus/dbus-message-handler.c: use atomic reference counting to
3755 reduce number of locks slightly; the global lock in here sucks
3757 * dbus/dbus-connection.c
3758 (_dbus_connection_update_dispatch_status_and_unlock): variant of
3759 update_dispatch_status that can be called with lock held; then use
3760 in a couple places to reduce locking/unlocking
3761 (dbus_connection_send): hold the lock over the whole function
3762 instead of acquiring it twice.
3764 * dbus/dbus-timeout.c (_dbus_timeout_new): handle OOM
3766 * bus/connection.c (bus_connections_setup_connection): fix access
3767 to already-freed memory.
3769 * dbus/dbus-connection.c: keep a little cache of linked list
3770 nodes, to avoid using the global linked list alloc lock in the
3771 normal send-message case. Instead we just use the connection lock
3772 that we already have to take.
3774 * dbus/dbus-list.c (_dbus_list_find_last): new function
3776 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec):
3777 change to use a struct for the atomic type; fix docs,
3778 they return value before increment, not after increment.
3780 * dbus/dbus-string.c (_dbus_string_append_4_aligned)
3781 (_dbus_string_append_8_aligned): new functions to try to
3782 microoptimize this operation.
3783 (reallocate_for_length): break this out of set_length(), to
3784 improve profile info, and also so we can consider inlining the
3787 * dbus/dbus-message.c (dbus_message_new_empty_header): init data
3788 strings with some preallocation, cuts down on our calls to realloc
3789 a fair bit. Though if we can get the "move entire string to empty
3790 string" optimization below to kick in here, it would be better.
3792 * dbus/dbus-string.c (_dbus_string_move): just call
3793 _dbus_string_move_len
3794 (_dbus_string_move_len): add a special case for moving
3795 an entire string into an empty string; we can just
3796 swap the string data instead of doing any reallocs.
3797 (_dbus_string_init_preallocated): new function
3799 2003-05-11 Havoc Pennington <hp@pobox.com>
3801 Write a "test-profile" that does echo client-server with threads;
3802 profile reveals lock contention, memcpy/realloc of buffers, and
3803 UTF-8 validation as hot spots. 20% of lock contention eliminated
3804 with dbus_atomic_inc/dec implementation on x86. Much remaining
3805 contention is global mempool locks for GList and DBusList.
3807 * dbus/dbus-sysdeps.c (_dbus_atomic_inc, _dbus_atomic_dec): add
3810 * dbus/dbus-connection.c (struct DBusConnection): use
3811 dbus_atomic_t for the reference count
3813 * dbus/dbus-message.c (struct DBusMessage): declare
3814 dbus_atomic_t values as volatile
3816 * configure.in: code to detect ability to use atomic integer
3817 operations in assembly, from GLib patch
3819 * dbus/dbus-internals.c (_dbus_verbose_real): call getpid every
3820 time, tired of it being wrong in threads and forked processes
3822 * glib/test-profile.c: a little program to bounce messages back
3823 and forth between threads and eat CPU
3825 * dbus/dbus-connection.c: add debug spew macros for debugging
3826 thread locks; include config.h at top; fix deadlock in
3827 dbus_connection_flush()
3829 2003-05-08 Havoc Pennington <hp@pobox.com>
3831 * dbus/dbus-spawn.c: s/_exit/exit/ because it was keeping gcov
3832 data from getting written, and there wasn't a good reason to
3835 * test/decode-gcov.c (mark_inside_dbus_build_tests): don't count
3836 dbus_verbose lines in test coverage
3837 (main): add list of functions sorted by # of untested blocks
3838 to the coverage report
3840 * dbus/dbus-mempool.c: put some test-only code in DBUS_BUILD_TESTS
3842 * dbus/dbus-marshal.c (_dbus_marshal_test): extend test coverage
3844 * dbus/dbus-message-handler.c (_dbus_message_handler_test):
3845 extend test coverage
3847 * test/data/auth/cancel.auth-script: test canceling an
3850 * dbus/Makefile.am: remove dbus-server-debug.[hc] for now, as they
3851 aren't used. in CVS history if we end up needing them.
3853 2003-05-04 Havoc Pennington <hp@pobox.com>
3855 * dbus/dbus-message-handler.c (_dbus_message_handler_test): add
3858 * dbus/dbus-marshal.c (_dbus_demarshal_string_array): fix this
3859 function, which assumed length was in # of strings, not bytes
3861 * dbus/dbus-message.c (_dbus_message_test): add tests for some
3864 * dbus/dbus-connection.c
3865 (_dbus_connection_queue_received_message): disable function for
3866 now, we are only using it in test mode
3868 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
3869 remove a mistaken FIXME
3871 2003-05-04 Havoc Pennington <hp@pobox.com>
3873 * dbus/dbus-connection.c (dbus_connection_preallocate_send):
3874 unlock mutex on successful return, patch from Anders Gustafsson
3876 2003-05-04 Havoc Pennington <hp@pobox.com>
3878 * dbus-glib-1.pc.in (Requires): fix dependencies, from
3881 2003-05-04 Havoc Pennington <hp@pobox.com>
3883 * tools/dbus-launch.c: implement
3885 * bus/main.c (main), bus/bus.c (bus_context_new):
3886 implement --print-pid and --fork
3888 2003-05-03 Havoc Pennington <hp@redhat.com>
3890 * dbus/dbus-address.c (dbus_parse_address): fix bug when a key in
3891 the address had no value, and add to test suite. Fix and
3892 regression test from Miloslav Trmac
3894 2003-05-03 Havoc Pennington <hp@pobox.com>
3896 * dbus/dbus-watch.c (dbus_watch_handle): warn and return if a
3897 watch is invalid when handled
3899 * tools/Makefile.am, tools/dbus-launch.c, tools/dbus-launch.1: add
3900 dbus-launch utility to launch the bus from a shell script. Didn't
3901 actually implement dbus-launch yet, it's just a placeholder still.
3903 2003-05-03 Havoc Pennington <hp@pobox.com>
3905 * bus/Makefile.am, bus/dbus-daemon-1.1.in: man page for the
3906 daemon; also documents daemon config file, so replaces
3907 doc/config-file.txt. Corrected some stuff from config-file.txt in
3908 the process of moving it.
3910 2003-05-03 Havoc Pennington <hp@pobox.com>
3912 * tools/Makefile.am, tools/dbus-send.1, tools/dbus-monitor.1:
3915 2003-05-03 Colin Walters <walters@verbum.org>
3917 * dbus/dbus-sysdeps.c (fill_user_info): Test against
3918 DBUS_UID_UNSET to determine whether to do a uid lookup or not.
3920 * Makefile.am: Update to use new .pc versioning scheme.
3922 2003-05-02 Havoc Pennington <hp@redhat.com>
3924 * bus/system.conf.in: allow send/receive to/from message bus
3927 2003-04-30 Havoc Pennington <hp@redhat.com>
3929 * configure.in: print a note when building with unit tests and
3932 2003-04-30 Havoc Pennington <hp@redhat.com>
3934 * Makefile.am: add a check-local that complains if you didn't
3935 configure with --enable-tests
3937 2003-04-29 Havoc Pennington <hp@redhat.com>
3939 * glib/dbus-gmain.c: docs cleanups
3941 * dbus/dbus-types.h: add docs on int64 types
3943 * dbus/dbus-memory.c: fix docs to avoid putting private API in
3944 public API docs section
3946 2003-04-29 Havoc Pennington <hp@redhat.com>
3948 * dbus-1.pc.in, dbus-glib-1.pc.in: rename these from
3949 dbus-1.0.pc.in, dbus-glib-1.0.pc.in. As these change with the
3950 parallel install API version, not with the D-BUS package version.
3952 * HACKING: move some of README over here
3954 * README: updates, and document API/ABI policy
3956 * configure.in: reindentation
3958 2003-04-29 Havoc Pennington <hp@redhat.com>
3960 * dbus/dbus.h: add "you have to define DBUS_API_SUBJECT_TO_CHANGE
3961 to use this library" to be sure people have the right
3964 2003-04-28 Havoc Pennington <hp@redhat.com>
3966 * configure.in: add --enable-docs which by default is auto yes if
3967 doxygen and db2html found, no otherwise; but can be forced on/off
3969 * doc/Makefile.am: conditionalize whether to build docs on
3972 2003-04-28 Havoc Pennington <hp@redhat.com>
3974 * configure.in: 0.10
3978 * bus/system.conf.in: add <includedir>system.d</includedir>
3980 * dbus/dbus-userdb.c (_dbus_user_database_lookup): fix bug when
3981 username was provided but not uid
3983 * bus/config-parser.c (struct BusConfigParser): keep track of
3984 whether the parser is toplevel or was included; change some
3985 of the error handling if it's included.
3987 2003-04-27 Havoc Pennington <hp@pobox.com>
3991 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
3992 report correct status if we finish processing authentication
3993 inside this function.
3995 * bus/activation.c (try_send_activation_failure): use
3996 bus_transaction_send_error_reply
3998 * bus/connection.c (bus_connection_get_groups): return an error
3999 explaining the problem
4001 * bus/bus.c (bus_context_check_security_policy): implement
4002 restriction here that inactive connections can only send the
4003 hello message. Also, allow bus driver to send anything to
4006 * bus/connection.c (bus_connection_complete): create the
4007 BusClientPolicy here instead of on-demand.
4008 (bus_connection_get_policy): don't return an error
4010 * dbus/dbus-message.c (dbus_message_new_error_reply): allow NULL
4011 sender field in message being replied to
4013 * bus/bus.c (bus_context_check_security_policy): fix silly typo
4014 causing it to return FALSE always
4016 * bus/policy.c (bus_client_policy_check_can_send): fix bug where
4017 we checked sender rather than destination
4019 2003-04-25 Havoc Pennington <hp@redhat.com>
4021 test suite is slightly hosed at the moment, will fix soon
4023 * bus/connection.c (bus_connections_expire_incomplete): fix to
4024 properly disable the timeout when required
4025 (bus_connection_set_name): check whether we can remove incomplete
4026 connections timeout after we complete each connection.
4028 * dbus/dbus-mainloop.c (check_timeout): fix this up a bit,
4029 probably still broken.
4031 * bus/services.c (bus_registry_acquire_service): implement max
4032 number of services owned, and honor allow/deny rules on which
4033 services a connection can own.
4035 * bus/connection.c (bus_connection_get_policy): report errors here
4037 * bus/activation.c: implement limit on number of pending
4040 2003-04-25 Havoc Pennington <hp@redhat.com>
4042 * dbus/dbus-transport.c (_dbus_transport_get_unix_user): fix bug
4043 where we used >= 0 instead of != DBUS_UID_UNSET.
4045 2003-04-25 Havoc Pennington <hp@redhat.com>
4047 * glib/dbus-gmain.c (remove_watch): fix for a crash when watches
4048 were toggled without add/remove, fix from Anders Gustafsson
4050 2003-04-24 Havoc Pennington <hp@redhat.com>
4052 * test/data/valid-config-files/basic.conf: add <limit> tags to
4055 * bus/config-parser.h, bus/config-parser.c, bus/bus.c: Implement
4056 <limit> tag in configuration file.
4058 2003-04-24 Havoc Pennington <hp@redhat.com>
4060 * bus/dispatch.c: somehow missed some name_is
4062 * dbus/dbus-timeout.c (_dbus_timeout_set_enabled)
4063 (_dbus_timeout_set_interval): new
4065 * bus/connection.c (bus_connections_setup_connection): record time
4066 when each connection is first set up, and expire them after the
4067 auth timeout passes.
4069 2003-04-24 Havoc Pennington <hp@redhat.com>
4071 * dbus/dbus-message.c (dbus_message_name_is): rename
4072 (dbus_message_service_is): rename
4073 (dbus_message_sender_is): rename
4074 (dbus_message_get_service): rename
4076 2003-04-24 Havoc Pennington <hp@redhat.com>
4078 * configure.in: add --enable-checks
4080 * dbus/dbus-message.c (dbus_message_new): reverse name/service arguments
4082 * dbus/dbus-connection.c (dbus_connection_preallocate_send): fix
4083 to use thread locks.
4084 (_dbus_connection_handler_destroyed_locked): move some private
4085 functions into proper docs group
4087 * dbus/dbus-internals.h: add _dbus_return_if_fail,
4088 _dbus_return_val_if_fail
4090 Throughout: use dbus_return_if_fail
4092 2003-04-23 James Willcox <jwillcox@gnome.org>
4095 * glib/dbus-gmain.c: (add_timeout), (wakeup_main), (create_source),
4096 (dbus_connection_setup_with_g_main),
4097 (dbus_server_setup_with_g_main):
4098 * glib/test-dbus-glib.c: (main):
4099 * glib/test-thread-client.c: (main):
4100 * glib/test-thread-server.c: (new_connection_callback), (main):
4101 * tools/dbus-monitor.c: (main):
4103 Added a GMainContext argument to dbus_connection_setup_with_g_main()
4104 and dbus_server_setup_with_g_main().
4106 2003-04-20 Havoc Pennington <hp@pobox.com>
4108 * doc/dbus-specification.sgml: document the restrictions on
4109 message and service names
4111 2003-04-22 Havoc Pennington <hp@redhat.com>
4113 * dbus/dbus-message.c, dbus/dbus-marshal.c: add 64-bit integer
4114 support, and do some code cleanups to share more code and
4115 speed up array marshal/demarshal.
4117 * dbus-1.0.pc.in (Cflags): put libdir include file in cflags
4119 * configure.in: generate dbus-arch-deps.h
4121 * dbus/dbus-protocol.h (DBUS_TYPE_INT64, DBUS_TYPE_UINT64): add
4124 2003-04-22 Havoc Pennington <hp@redhat.com>
4126 * test/data/valid-messages/opposite-endian.message: fix test
4127 to use proper type for rply field
4129 * test/data/invalid-messages: add tests for below validation
4131 * dbus/dbus-message.c (decode_header_data): validate field types,
4132 and validate that named fields are valid names
4133 (decode_name_field): consider messages in the
4134 org.freedesktop.Local. namespace to be invalid.
4136 * dbus/dbus-string.c (_dbus_string_validate_name): new
4138 2003-04-19 Havoc Pennington <hp@pobox.com>
4140 * bus/driver.c (bus_driver_handle_hello): check limits and
4141 return an error if they are exceeded.
4143 * bus/connection.c: maintain separate lists of active and inactive
4144 connections, and a count of each. Maintain count of completed
4145 connections per user. Implement code to check connection limits.
4147 * dbus/dbus-list.c (_dbus_list_unlink): export
4149 * bus/bus.c (bus_context_check_security_policy): enforce a maximum
4150 number of bytes in the message queue for a connection
4152 2003-04-18 Havoc Pennington <hp@pobox.com>
4154 * dbus/dbus-auth.c (record_mechanisms): memleak fixes
4156 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): fix some
4159 * dbus/dbus-keyring.c (add_new_key): fix a memleak, and
4160 on realloc be sure to update the pointer in the keyring
4162 * dbus/dbus-string.c (_dbus_string_zero): compensate for align
4163 offset to avoid writing to unallocated memory
4165 * dbus/dbus-auth.c (process_rejected): return FALSE if we fail to
4166 try the next mechanism, so we properly handle OOM
4168 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): fix double-free
4170 (_dbus_keyring_new): fix OOM bug
4171 (_dbus_keyring_new_homedir): always set error; impose a maximum
4172 number of keys we'll load from the file, mostly to speed up the
4173 test suite and make its OOM checks more useful, but also for
4176 * dbus/dbus-auth.c (process_error_server): reject authentication
4177 if we get an error from the client
4178 (process_cancel): on cancel, send REJECTED, per the spec
4179 (process_error_client): send CANCEL if we get an error from the
4182 2003-04-18 Havoc Pennington <hp@pobox.com>
4184 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix UMR in verbose
4187 * dbus/dbus-auth.c (handle_client_data_cookie_sha1_mech): fix OOM
4190 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): only whine
4191 about DBUS_TEST_HOMEDIR once
4193 * bus/Makefile.am (TESTS_ENVIRONMENT): put DBUS_TEST_HOMEDIR in
4196 * bus/dispatch.c (bus_dispatch_sha1_test): actually load sha1
4197 config file so we test the right thing
4199 Throughout: assorted docs improvements
4201 2003-04-18 Havoc Pennington <hp@pobox.com>
4203 * glib/dbus-gmain.c: adapt to watch changes
4205 * bus/bus.c, bus/activation.c, etc.: adjust to watch changes
4207 * dbus/dbus-server.h: remove dbus_server_handle_watch
4209 * dbus/dbus-connection.h: remove dbus_connection_handle_watch
4211 * dbus/dbus-watch.c (dbus_watch_handle): change DBusWatch to work
4212 like DBusTimeout, so we don't need dbus_connection_handle_watch
4215 2003-04-17 Havoc Pennington <hp@redhat.com>
4217 * dbus/dbus-userdb.c, dbus/dbus-sysdeps.c: redo all the passwd
4218 database usage so it all goes via the DBusUserDatabase cache.
4220 2003-04-17 Havoc Pennington <hp@redhat.com>
4222 * dbus/dbus-mainloop.c (_dbus_loop_iterate): fix logic so that if
4223 there was an OOM watch we skipped, we always return TRUE so we
4224 iterate again to have a look at it again. Fixes test suite hang.
4225 Code rearrangement also lets us lose some memset and only iterate
4226 over callbacks once.
4228 * bus/driver.c (bus_driver_handle_message): sense of test for
4231 2003-04-16 Havoc Pennington <hp@pobox.com>
4233 * doc/dbus-specification.sgml: make spec say serials are unsigned
4235 * dbus/dbus-message.h: change message serials to unsigned
4237 * dbus/dbus-connection.c: adapt to message serials being unsigned
4239 2003-04-15 Havoc Pennington <hp@pobox.com>
4241 * bus/bus.c: create and keep around a shared DBusUserDatabase
4244 * bus/connection.c (bus_connection_get_groups): don't cache
4245 groups for user in the connection object, since user database
4246 object now does that.
4248 2003-04-16 Havoc Pennington <hp@redhat.com>
4250 * dbus/dbus-message.c (_dbus_message_add_size_counter): keep a
4251 list of size counters
4252 (_dbus_message_loader_putback_message_link): put back a popped link
4254 * dbus/dbus-connection.c
4255 (dbus_connection_set_max_live_messages_size): rename
4257 (dbus_connection_get_outgoing_size): get size of outgoing
4259 (_dbus_connection_set_connection_counter): remove this cruft
4261 2003-04-14 Havoc Pennington <hp@redhat.com>
4263 * dbus/dbus-userdb.c: user database abstraction, mostly to get
4264 caching, but at some point we might want to be able to use a
4267 * bus/dispatch.c (bus_dispatch_sha1_test): add a test that uses
4268 SHA1 conf file to test the sha1 auth mechanism, since the regular
4269 test always uses EXTERNAL when available.
4272 test/data/valid-config-files/debug-allow-all-sha1.conf.in:
4273 add conf file that requires use of sha1 auth
4275 2003-04-13 Havoc Pennington <hp@pobox.com>
4277 * tools/dbus-send.c, tools/dbus-monitor.c: two utility programs
4278 from Philip Blundell to send messages and monitor them.
4280 2003-04-13 Havoc Pennington <hp@pobox.com>
4282 * dbus/dbus-mainloop.c: fix some reentrancy issues by refcounting
4285 * test/data/valid-config-files/debug-allow-all.conf.in: allow all
4288 * dbus/dbus-transport.c (_dbus_transport_get_dispatch_status):
4289 fix to only recover unused bytes if we're already authenticated
4290 (_dbus_transport_get_is_authenticated): fix to still mark us
4291 authenticated if there are unused bytes.
4293 * bus/dispatch.c: implement security policy checking
4295 * bus/connection.c (bus_transaction_send_from_driver): new
4297 * bus/bus.c (bus_context_check_security_policy): new
4299 * bus/dispatch.c (send_service_nonexistent_error): delete this,
4300 now we just set the DBusError and it gets converted to an error
4303 * bus/connection.c (allow_user_function): enable code using actual
4304 data from the config file
4306 * bus/policy.c (list_allows_user): handle wildcard rules for
4307 user/group connection perms
4309 2003-04-13 Havoc Pennington <hp@pobox.com>
4311 * bus/config-parser.c: Load up the BusPolicy and BusPolicyRules
4313 * dbus/dbus-sysdeps.c (_dbus_get_user_id): new function
4315 * bus/policy.c (bus_policy_append_mandatory_rule)
4316 (bus_policy_append_default_rule, bus_policy_append_user_rule)
4317 (bus_policy_append_group_rule): new functions
4319 2003-04-12 Havoc Pennington <hp@pobox.com>
4321 * bus/config-parser.c (bus_config_parser_new): fix a memleak
4323 * dbus/dbus-sysdeps.c: change DBusCredentials to use longs for
4324 the pid/gid/uid, just for paranoia.
4326 * test/break-loader.c (randomly_do_n_things): find a byte
4327 containing a type code, and randomly change it to a different
4330 2003-04-12 Havoc Pennington <hp@pobox.com>
4332 * bus/policy.h: change BusPolicy to be the thing from the config
4333 file, and rename old BusPolicy to BusClientPolicy
4335 * bus/bus.c, bus/connection.c, bus/config-parser.c: change to
4336 match change in how policy works
4338 * dbus/dbus-internals.h: mark assert_not_reached as
4339 __attribute((noreturn))__
4341 2003-04-11 Havoc Pennington <hp@redhat.com>
4343 * doc/dbus-specification.sgml: fix a spot with the wrong name for
4344 the broadcast service. Use boolean return for ServiceExists.
4346 2003-04-11 Havoc Pennington <hp@redhat.com>
4348 * configure.in: add another directory to look for qt in.
4350 2003-04-11 Havoc Pennington <hp@redhat.com>
4352 * AUTHORS: add Colin Walters
4354 2003-04-11 Havoc Pennington <hp@redhat.com>
4360 2003-04-11 Havoc Pennington <hp@redhat.com>
4362 * bus/messagebus.in: remove pid file when stopping the
4363 message bus, since the bus won't have privileges to remove it
4366 2003-04-11 Havoc Pennington <hp@redhat.com>
4368 * bus/bus.c (bus_context_new): move credentials change after
4371 2003-04-11 Havoc Pennington <hp@pobox.com>
4373 * test/decode-gcov.c: add "below average functions" to the
4374 coverage report, and change how some of the code works.
4376 * bus/test-main.c: bracket some stuff in DBUS_BUILD_TESTS so it's
4377 not in the coverage stats.
4379 * test/test-service.c (main): use _dbus_verbose not fprintf in a
4380 couple places so running the test suite doesn't result in megaspam.
4382 2003-04-11 Havoc Pennington <hp@pobox.com>
4384 * bus/dispatch.c (check_existent_service_activation): accept a no
4385 memory error in a place we didn't before
4387 * bus/test.c (bus_test_run_everything): remove hacky "do it twice
4388 in case the first one failed," since the test suite is less
4391 2003-04-10 Havoc Pennington <hp@pobox.com>
4393 * bus/dispatch.c (check_segfault_service_activation): add test
4394 for launching an executable that just crashes.
4396 * test/test-segfault.c (main): try setting coredumpsize to 0 so we
4397 don't leave a million cores. We'll see how portable this is.
4399 2003-04-10 Havoc Pennington <hp@pobox.com>
4401 * dbus/dbus-spawn.c (_dbus_spawn_async_with_babysitter): move all
4402 the possible parent failures before we fork, so that we don't
4403 fail to create a babysitter after creating the child.
4405 * bus/activation.c (bus_activation_activate_service): kill child
4406 if we don't successfully complete the activation.
4408 2003-04-10 Havoc Pennington <hp@redhat.com>
4410 * dbus/dbus-connection.c (dbus_connection_flush): don't spin on
4411 the connection if it's disconnected
4413 * bus/activation.c (bus_activation_service_created): use new
4414 transaction features to roll back removal of pending activation if
4415 we don't successfully create the service after all. Don't remove
4416 pending activation if the function fails.
4418 * dbus/dbus-list.c (_dbus_list_insert_before_link)
4419 (_dbus_list_insert_after_link): new code to facilitate
4422 * dbus/dbus-hash.c (_dbus_hash_table_insert_string_preallocated):
4423 new functionality, so we can preallocate the ability to insert
4426 * bus/connection.c (bus_transaction_add_cancel_hook): new function
4427 allowing us to put custom hooks in a transaction to be used for
4428 cancelling said transaction
4430 * doc/dbus-specification.sgml: add some discussion of secondary
4431 service owners, and disallow zero-length service names
4433 * bus/services.c (bus_registry_acquire_service): new function,
4434 splits out part of bus_driver_handle_acquire_service() and fixes
4435 a bug where we didn't remove the service doing the acquiring
4436 from the secondary queue if we failed to remove the current owner
4437 from the front of the queue.
4439 2003-04-10 Alexander Larsson <alexl@redhat.com>
4441 * doc/dbus-specification.sgml:
4442 s/org.freedesktop.Broadcast/org.freedesktop.DBus.Broadcast/
4444 2003-04-10 Alexander Larsson <alexl@redhat.com>
4449 Added files to cvsignore
4451 * dbus/dbus-message.h:
4452 * dbus/dbus-message.c: (dbus_message_iter_get_named):
4453 Make get_named() take two out argument and return a boolean.
4454 (dbus_message_iter_get_args_valist):
4455 Update usage of get_named().
4456 (dbus_message_iter_append_byte):
4458 (dbus_message_iter_append_named)
4460 (message_iter_test), (check_message_handling_type), (_dbus_message_test):
4463 2003-04-10 Alexander Larsson <alexl@redhat.com>
4465 * dbus/dbus-marshal.[ch]:
4466 Add array_type_pos argument to _dbus_marshal_validate_arg.
4467 Let you pass a NULL end_pos to _dbus_marshal_validate_type.
4469 * dbus/dbus-message.[ch]:
4470 Multi-dimensional arrays have full type specification in the
4471 outermost array. Iter code re-arranged to handle this.
4472 Added some more iter tests.
4474 * doc/dbus-specification.sgml:
4477 Update new array encoding description.
4478 Correct DBUS_SERVICE_FLAGS_REPLACE_EXISTING description.
4480 * test/data/invalid-messages/array-with-mixed-types.message:
4481 * test/data/valid-messages/array-of-array-of-uint32.message:
4482 Change to the new array format.
4484 * test/data/invalid-messages/too-short-dict.message:
4487 * test/data/valid-messages/recursive-types.message:
4488 Fix up and extend test.
4490 2003-04-10 Havoc Pennington <hp@pobox.com>
4492 * bus/dispatch.c: lots of fixes
4494 * dbus/dbus-mainloop.c (_dbus_loop_dispatch): export
4495 (_dbus_loop_iterate): remove old "quit if no callbacks" code,
4496 that was crack, broke the test service.
4498 * dbus/dbus-transport.c (_dbus_transport_open): fix error
4499 handling to avoid piling up errors if we get a failure on the
4502 * dbus/dbus-internals.c (_dbus_real_assert_not_reached): include
4503 pid in assertion failures.
4505 * dbus/dbus-mainloop.c (_dbus_loop_iterate): use static arrays up
4506 to some fixed size of file descriptor array. Don't return TRUE
4507 anytime a timeout exists, that led to lots of busy loop silliness
4510 2003-04-09 Havoc Pennington <hp@redhat.com>
4512 * dbus/dbus-mainloop.c (check_timeout): fix timeouts, I thought
4513 I'd checked this in earlier but hadn't.
4515 2003-04-09 Havoc Pennington <hp@redhat.com>
4517 * bus/dispatch.c (bus_dispatch_test): get a bit further through
4518 the activation test (man this is getting old!)
4520 2003-04-09 Havoc Pennington <hp@redhat.com>
4522 * test/test-utils.c: use dispatch status function to fix this up
4524 * bus/connection.c (connection_watch_callback): don't dispatch
4526 (connection_timeout_callback): don't dispatch from here
4527 (bus_connections_setup_connection): set the dispatch status function
4528 (bus_connection_disconnected): unset it
4530 * dbus/dbus-mainloop.c (_dbus_loop_queue_dispatch): new function
4531 used to add a connection to be dispatched
4532 (_dbus_loop_iterate): do the dispatching at the end of each
4535 * dbus/dbus-connection.c
4536 (dbus_connection_set_dispatch_status_function): new function
4537 allowing us to fix up main loop usage
4538 (_dbus_connection_last_unref): free all the various function
4540 (dbus_connection_dispatch): call the DispatchStatusFunction
4541 whenever this function returns
4542 (dbus_connection_handle_watch): call DispatchStatusFunction
4543 (dbus_connection_send_with_reply_and_block): call DispatchStatusFunction
4544 (reply_handler_timeout): call DispatchStatusFunction
4545 (dbus_connection_flush): call DispatchStatusFunction
4547 2003-04-09 Havoc Pennington <hp@redhat.com>
4549 * dbus/dbus-bus.c (dbus_bus_register): fix up error handling and
4552 * bus/dispatch.c (check_service_activated): fix bug in test
4554 * dbus/dbus-mainloop.c (check_timeout): fix this up
4556 * dbus/dbus-internals.c (_dbus_verbose_real): include PID in
4557 verbose output so we can sort out output from different processes,
4558 e.g. in the activation case.
4560 2003-04-08 Colin Walters <walters@gnu.org>
4562 * bus/bus.c (struct BusContext) [pidfile]: New member, to store
4564 (bus_context_new): Set it.
4565 (bus_context_unref): Use it to delete the pid file.
4567 2003-04-08 Havoc Pennington <hp@redhat.com>
4569 * test/data/invalid-messages/array-with-mixed-types.message:
4570 regression test that fails for the moment
4572 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): reorder
4573 tests for convenience
4575 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): don't allow
4576 array of nil, it broke things.
4578 * test/data/invalid-messages/array-of-nil.message: regression test
4580 * test/data/valid-messages/array-of-array-of-uint32.message:
4581 happened to write this so added it to suite
4583 2003-04-08 Havoc Pennington <hp@redhat.com>
4585 * bus/driver.c (bus_driver_handle_acquire_service): init
4586 retval/reply before checking name
4588 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add a
4589 recursion depth argument
4591 * dbus/dbus-message.h (struct DBusMessageIter): put some padding
4592 in the public struct for future extension
4594 * dbus/dbus-message-builder.c (_dbus_message_data_load): fix
4597 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): fix a verbose
4600 * doc/dbus-specification.sgml: fix typo
4602 2003-04-08 Alexander Larsson <alexl@redhat.com>
4604 Implemented recursive types, named types and new-style iters
4607 * glib/test-thread-client.c: (thread_func):
4608 * glib/test-thread-server.c: (handle_test_message):
4609 * test/test-service.c: (handle_echo):
4618 * dbus/dbus-internals.c: (_dbus_type_to_string):
4619 Update for new types.
4621 * dbus/dbus-marshal.[ch]:
4622 Implement recursive types and the new marshalling format.
4623 Remove hardcoded dict marshalling.
4624 Marshal named types.
4626 * dbus/dbus-message-builder.c:
4628 Remove references to old types
4630 * dbus/dbus-message.[ch]:
4631 New non-refcounted iter API that supports recursive iters.
4632 Use iters for appending, including support for recursive
4634 Add byte and named type support.
4635 Update everything to new marshalling formats.
4636 Add tests for new API.
4638 * dbus/dbus-protocol.h:
4639 Remove old array types.
4640 Add types: BYTE, ARRAY, DICT, NAMED
4642 * dbus/dbus-string.c:
4643 * dbus/dbus-sysdeps.c:
4644 Make parse_double locale safe.
4646 * dbus/dbus-test-main.c:
4652 * doc/dbus-specification.sgml:
4655 * test/data/incomplete-messages/missing-body.message:
4656 * test/data/invalid-messages/bad-boolean.message:
4657 * test/data/invalid-messages/bad-boolean-array.message:
4658 * test/data/invalid-messages/boolean-array-length-too-long.message-raw:
4659 * test/data/invalid-messages/boolean-has-no-value.message-raw:
4660 * test/data/invalid-messages/too-short-dict.message:
4661 * test/data/valid-messages/dict-simple.message:
4662 * test/data/valid-messages/dict.message:
4663 * test/data/valid-messages/emptiness.message:
4664 * test/data/valid-messages/lots-of-arguments.message:
4665 * test/data/valid-messages/no-padding.message:
4666 * test/data/valid-messages/recursive-types.message:
4667 Add missing NAME fields
4668 Fix up dicts & arrays
4670 * test/data/invalid-messages/dict-with-nil-value.message:
4671 Removed, this is not invalid anymore.
4673 * test/data/valid-messages/recursive-types.message:
4674 Add new test for deeply recursive types.
4676 2003-04-07 Havoc Pennington <hp@pobox.com>
4678 * bus/driver.c (bus_driver_handle_acquire_service): return an
4679 error if you try to acquire a service that starts with ':'
4681 2003-04-07 Havoc Pennington <hp@redhat.com>
4683 * doc/dbus-specification.sgml: require that base service names
4684 start with ':' and that the base service is created/deleted
4685 as first and last things a connection does on the bus
4687 * bus/dispatch.c (check_existent_service_activation): lots more
4688 work on the activation test; it doesn't fully pass yet...
4690 * test/test-service.c (main): fix so we don't memleak the
4691 connection to the message bus
4692 (filter_func): accept a message asking us to exit
4694 2003-04-06 Havoc Pennington <hp@pobox.com>
4696 * qt/Makefile.am (dbusinclude_HEADERS): install dbus-qt.h,
4699 * configure.in: fixes to Qt detection from Colin Walters
4701 * doc/Makefile.am: Only remove generated docbook dirs if they
4702 exist, from Colin Walters
4704 * dbus/dbus-bus.c: change how we set well-known connections to
4705 NULL, so that it works if a single connection is stored in
4706 two well-known array slots.
4708 * test/Makefile.am: remove a lot of stuff that isn't immediately
4709 useful, it's in CVS history if we want it.
4711 * test/test-service.c: use dbus-mainloop instead of that
4714 2003-04-06 Havoc Pennington <hp@pobox.com>
4716 * dbus/Makefile.am: split lists of sources into stuff that goes in
4717 the library, util functions that go in the lib and are also used
4718 elsewhere, and util functions that are used in tests/daemon but
4719 don't go in the lib.
4721 * dbus/dbus-mainloop.h, dbus/dbus-mainloop.c: move bus/loop.[hc]
4722 here so it can be used in test binaries also
4724 2003-04-06 Havoc Pennington <hp@pobox.com>
4726 * dbus/dbus-sysdeps.c (_dbus_become_daemon): write the pidfile
4727 here in the parent process, so we can return an error if it
4728 fails. Also, move some of the code into the child so the parent
4729 is less hosed if we fail midway through.
4731 * bus/bus.c (bus_context_new): move pidfile detection further up
4732 in the function, before we start overwriting sockets and such.
4734 * bus/messagebus.in: adjust this a bit, not sure if it will work.
4736 * configure.in: add --with-system-pid-file and --with-system-socket
4738 2003-04-06 Colin Walters <walters@verbum.org>
4740 * configure.in (DBUS_SYSTEM_PID_FILE): New variable.
4742 * bus/system.conf.in: Declare a pidfile.
4744 * bus/bus.c (bus_context_new): Test for an existing pid file, and
4745 create one (if appropriate).
4747 * bus/config-parser.c (enum ElementType) [ELEMENT_PIDFILE]: New.
4748 (struct BusConfigParser) [pidfile]: New.
4749 (element_type_to_name, merge_included, start_busconfig_child)
4750 (bus_config_parser_end_element, bus_config_parser_content): Handle it.
4751 (bus_config_parser_unref): Free it.
4752 (bus_config_parser_get_pidfile): New function.
4754 * bus/config-parser.h (_dbus_write_pid_file): Prototype.
4756 * dbus/dbus-errors.h (DBUS_ERROR_PIDFILE_EXISTS): New error.
4758 * dbus/dbus-sysdeps.c (_dbus_write_pid_file): New function.
4760 * dbus/dbus-sysdeps.h: Prototype it.
4762 2003-04-06 Havoc Pennington <hp@pobox.com>
4764 * bus/bus.c (bus_context_new): print the address in here, rather
4765 than in main(), because we need to do it before forking the daemon
4767 * bus/dispatch.c (send_service_nonexistent_error): set the sender
4768 on the service nonexistent error
4770 * bus/driver.c (bus_driver_handle_acquire_service): set the
4771 sender on the AcquireService reply
4773 * test/data/valid-config-files/debug-allow-all.conf.in: Make test
4774 server also listen on a UNIX socket so services can connect to it.
4776 2003-04-06 Havoc Pennington <hp@pobox.com>
4778 * dbus/dbus-threads.c: Redo how the fake debug mutexes are done
4779 so it detects deadlocks and also we actually init threads when
4782 2003-04-06 Havoc Pennington <hp@pobox.com>
4784 * dbus/dbus-server-unix.c (_dbus_server_new_for_domain_socket):
4785 save the domain socket name, and unlink it when we disconnect the
4786 server. Means that at least when we exit normally, we won't leave
4787 a bunch of junk in /tmp
4789 * dbus/dbus-transport-unix.c
4790 (_dbus_transport_new_for_domain_socket): code cleanup (nicer
4791 memory management). (I was making a real change here but then
4794 2003-04-06 Havoc Pennington <hp@pobox.com>
4796 * bus/bus.c (bus_context_new): fix wrong handling of
4797 server_data_slot_unref() in the error case.
4799 * dbus/dbus-internals.h (_dbus_assert): change so it passes
4800 "(condition) != 0" to _dbus_real_assert so that
4801 "_dbus_assert (pointer)" doesn't cause a warning
4803 * bus/main.c (main): accept --print-address option to print out
4804 the message bus address
4806 * dbus/dbus-sysdeps.c (_dbus_generate_random_ascii): export this
4808 * dbus/dbus-transport.c (_dbus_transport_open): special error for
4809 "tmpdir" option to unix: address on client side
4811 * dbus/dbus-server.c (dbus_server_listen): handle "tmpdir" option
4814 * configure.in (TEST_SOCKET_DIR): locate a temporary directory
4815 we can use to create sockets in the test suite.
4817 * bus/main.c (signal_handler): on SIGTERM, exit the daemon
4818 cleanly. To be used for testing.
4820 * dbus/dbus-spawn.c (babysit): use _dbus_set_signal_handler()
4822 * dbus/dbus-sysdeps.c (_dbus_set_signal_handler): new
4824 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
4825 handle trying to call this when there's no servers active
4827 2003-04-05 Havoc Pennington <hp@pobox.com>
4833 2003-04-05 Havoc Pennington <hp@pobox.com>
4835 * bus/bus.c (setup_server): fix this so dbus-daemon-1 doesn't
4836 crash on startup. Need to get "try starting the daemon"
4837 in the test suite I guess. ;-)
4839 * dbus/dbus-server.h, dbus/dbus-server.c: remove the stuff that
4840 tracked the number of open connections; it's better done in
4841 application-specific code as you want it to span all servers etc.
4843 2003-04-05 Havoc Pennington <hp@pobox.com>
4845 * bus/Makefile.am (install-data-hook): add missing DESTDIR,
4846 patch from Colin Walters
4848 2003-04-05 Havoc Pennington <hp@pobox.com>
4850 * doc/config-file.txt (Elements): fix docs of <auth> to reflect
4851 reality; in fact multiple mechanisms are allowed.
4853 * dbus/dbus-internals.c (_dbus_real_assert)
4854 (_dbus_real_assert_not_reached): move guts of _dbus_assert() and
4855 _dbus_assert_not_reached() into functions, so that they don't show
4856 up in basic block counts for test coverage, and don't use up as
4857 much disk space. Does mean slower execution speed though, so
4858 assumes --disable-asserts is the normal production case.
4860 2003-04-05 Havoc Pennington <hp@pobox.com>
4862 * test/Makefile.am (dist-hook): also dist *.in files
4868 2003-04-05 Havoc Pennington <hp@pobox.com>
4870 * dbus/dbus-string.c: docs warning
4872 * dbus/dbus-spawn.c: missing docs
4874 * dbus/dbus-memory.c (struct ShutdownClosure): missing docs
4876 2003-04-05 Havoc Pennington <hp@pobox.com>
4878 * bus/loop.c (bus_loop_iterate): fix the timeout code, using
4881 * dbus/dbus-spawn.c (_dbus_babysitter_unref): set sitter_pid
4882 to -1 once we've reaped the babysitter
4883 (_dbus_babysitter_handle_watch): do as much work as we can, not
4886 * bus/activation.c: add code using DBusBabysitter so that we
4887 handle it when a service fails to start up properly.
4888 (bus_activation_service_created): don't remove the activation
4889 entries as we go, just let them get removed when we free the pending
4890 activation. Unref reply messages after sending them.
4892 2003-04-05 Havoc Pennington <hp@pobox.com>
4894 * test/decode-gcov.c (main): print per-directory stats in the report
4896 * Makefile.am (coverage-report.txt): don't include test/* in gcov stats
4898 2003-04-05 Havoc Pennington <hp@pobox.com>
4900 * Makefile.am (coverage-report.txt): add target "coverage-report.txt"
4902 * test/decode-gcov.c: hack up a little program to suck data
4903 out of gcov files. Yes this is sort of silly.
4905 * configure.in: define something in config.h and do an
4906 AM_CONDITIONAL when gcov is enabled
4908 2003-04-04 Havoc Pennington <hp@redhat.com>
4910 * dbus/dbus-spawn.c, dbus/dbus-spawn.h: Change dbus_spawn to
4911 return a "babysitter" object that is used to monitor the status of
4912 the spawned process and reap it when required.
4914 * test/test-segfault.c, test/test-exit.c,
4915 test/test-sleep-forever.c: binaries that do various lame things,
4916 used in the test suite.
4918 * dbus/dbus-sysdeps.c: kill _dbus_errno_to_string()
4920 2003-04-03 Havoc Pennington <hp@pobox.com>
4922 * dbus/dbus-spawn.c: Move dbus-spawn into a separate file
4923 in preparation for modifying it, dbus-sysdeps is getting
4926 2003-04-03 Havoc Pennington <hp@redhat.com>
4928 * bus/loop.h, bus/loop.c: make the mainloop an object so we can
4931 * bus/*.[hc]: adapt to mainloop change
4933 2003-04-03 Havoc Pennington <hp@redhat.com>
4935 * bus/activation.c (load_directory): fix up memleaks
4936 (bus_activation_entry_free): free the entry
4938 * dbus/dbus-bus.c (dbus_bus_acquire_service): return an error if
4939 we get one from the message bus; fix memleaks.
4941 * dbus/dbus-message.c (dbus_set_error_from_message): new function
4943 2003-04-03 Havoc Pennington <hp@pobox.com>
4945 * bus/config-parser.c (bus_config_parser_unref): free
4946 list of mechanisms, bug discovered by test suite enhancements
4947 (putting system.conf and session.conf into suite)
4949 * test/Makefile.am, test/test-service.c: add placeholder for a
4950 test service that we'll activate as part of test suite. Doesn't
4953 * dbus/dbus-sysdeps.c (_dbus_setenv): support unsetenv by
4954 setting NULL value, and use system malloc not dbus_malloc()
4955 when we have unavoidable memleakage.
4957 * dbus/dbus-bus.c (dbus_bus_get): fix bug where bus type of 0
4958 didn't work, and support DBUS_BUS_ACTIVATION.
4960 * bus/activation.c (child_setup): pass our well-known bus type to
4963 * bus/config-parser.c: support <type> to specify well-known type
4965 * doc/dbus-specification.sgml: document the env variables to
4966 locate well-known buses and find service activator
4968 2003-04-02 Havoc Pennington <hp@redhat.com>
4970 * test/Makefile.am (all-local): add a rule to copy tests to
4971 builddir, so we can have generated tests. Use this to remove the
4972 silly hack for testing system.conf and session.conf. Will use this
4973 shortly to generate .service files pointing to test binaries.
4975 2003-04-02 Havoc Pennington <hp@redhat.com>
4977 * dbus/dbus-string.c (set_length): fix a bug - we allocated max of
4978 current alloc and needed new length, not max of the doubled
4979 allocation and needed new length. Also, when building tests,
4980 don't do the double-allocation stuff, just realloc every time.
4982 2003-04-02 Havoc Pennington <hp@redhat.com>
4984 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): include filenames
4986 (_dbus_string_get_dirname): new
4987 (_dbus_sysdeps_test): new
4988 (_dbus_directory_open): include dirnames in error messages
4990 * bus/config-parser.c: interpret <include> and <includedir> and
4991 <servicedir> relative to config file location if the given
4992 filename is not absolute.
4994 * dbus/dbus-string.c (_dbus_string_find_byte_backward): new
4996 2003-04-02 Havoc Pennington <hp@redhat.com>
4998 * bus/connection.c (bus_transaction_send_error_reply): set sender
4999 service for the error, and unref the reply on success
5001 * bus/activation.c: convert to use BusTransaction so OOM can be
5003 (bus_activation_service_created): set sender of the message
5005 2003-04-01 Havoc Pennington <hp@redhat.com>
5007 * bus/config-parser.c, bus/bus.c: implement <servicedir> and
5008 <includedir> (at least mostly)
5010 * dbus/dbus-sysdeps.c (_dbus_change_identity): set the group ID
5011 first, then the user ID
5013 2003-04-01 Havoc Pennington <hp@pobox.com>
5015 * dbus/dbus-server.c (dbus_server_set_auth_mechanisms): new
5018 * dbus/dbus-auth.c (_dbus_auth_set_mechanisms): new
5020 * dbus/dbus-internals.c (_dbus_dup_string_array): new function
5022 * dbus/dbus-sysdeps.c (_dbus_listen_unix_socket): chmod the
5023 socket 0777, and unlink any existing socket.
5025 * bus/bus.c (bus_context_new): change our UID/GID and fork if
5026 the configuration file so specifies; set up auth mechanism
5029 * bus/config-parser.c (bus_config_parser_content): add support
5030 for <fork> option and fill in code for <auth>
5032 * bus/system.conf.in: add <fork/> to default configuration,
5033 and limit auth mechanisms to EXTERNAL
5035 * doc/config-file.txt (Elements): add <fork>
5037 * dbus/dbus-sysdeps.c (_dbus_become_daemon): new function
5038 (_dbus_change_identity): new function
5040 2003-03-31 Havoc Pennington <hp@redhat.com>
5042 * dbus/dbus-sysdeps.c (_dbus_connect_unix_socket)
5043 (_dbus_listen_unix_socket): fix off-by-one error in null
5044 termination spotted by Nalin
5046 2003-03-31 Havoc Pennington <hp@redhat.com>
5048 * dbus/dbus-keyring.c (_dbus_keyring_new_homedir): allow setting
5049 DBUS_TEST_HOMEDIR when tests are enabled, so we can test without
5050 having a real home directory available.
5052 2003-03-31 Havoc Pennington <hp@redhat.com>
5054 * bus/Makefile.am (install-data-hook): create /var/run/dbus
5056 * bus/messagebus.in: add init script for Red Hat /etc/init.d
5058 * configure.in: add support for specifying a style of init script
5061 2003-03-31 Havoc Pennington <hp@redhat.com>
5063 Fix some annoying DBusString API and fix all affected code.
5065 * dbus/dbus-string.c (_dbus_string_init): get rid of annoying
5067 (_dbus_string_get_data): change to return string instead of using
5069 (_dbus_string_get_const_data): ditto
5070 (_dbus_string_get_data_len): ditto
5071 (_dbus_string_get_const_data_len): ditto
5073 2003-03-31 Havoc Pennington <hp@redhat.com>
5075 * bus/main.c (main): fix up the command line arguments to be nicer
5077 2003-03-31 Havoc Pennington <hp@redhat.com>
5079 * dbus/Makefile.am (INCLUDES): use EXPANDED_LOCALSTATEDIR to
5080 define DBUS_SYSTEM_BUS_PATH as we want to compile in the same
5081 final location that lands in the config file
5083 * bus/config-loader-expat.c (bus_config_load): fix type of
5086 * doc/TODO: remove TODO item for dbus_bus_get()
5088 * dbus/dbus-bus.c (bus_data_free): add missing lock/unlock
5090 2003-03-31 Havoc Pennington <hp@pobox.com>
5092 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_domain_socket)
5093 (_dbus_transport_new_for_tcp_socket): these didn't need the "server"
5094 argument since they are always client side
5096 * dbus/dbus-server.c (dbus_server_get_address): new function
5098 * bus/main.c (main): take the configuration file as an argument.
5100 * test/data/valid-config-files/debug-allow-all.conf: new file to
5101 use with dispatch.c tests for example
5103 * bus/test-main.c (main): require test data dir
5105 * bus/bus.c (bus_context_new): change this to take a
5106 configuration file name as argument
5108 * doc/config-file.txt (Elements): add <servicedir>
5110 * bus/system.conf, bus/session.conf: new files
5112 * dbus/dbus-bus.c (dbus_bus_get): look for system bus on
5113 well-known socket if none set
5115 * configure.in: create system.conf and session.conf
5117 2003-03-30 Havoc Pennington <hp@pobox.com>
5119 * bus/config-parser.c: hacking
5121 * dbus/dbus-memory.c: don't use DBusList for the list of stuff
5122 to shut down, since it could cause weirdness with the DBusList
5125 * dbus/dbus-list.c (_dbus_list_test): add tests for the
5126 link-oriented stack routines
5127 (alloc_link): free the mempool if the first alloc from it fails
5129 * dbus/dbus-mempool.c (struct DBusMemBlock): fix alignment issue
5131 * dbus/dbus-string.c (UNICODE_VALID): sync new version of this
5133 (_dbus_string_skip_white): new
5135 * doc/config-file.txt (Elements): add <includedir>
5137 2003-03-28 Havoc Pennington <hp@pobox.com>
5139 * dbus/dbus-string.c (_dbus_string_copy_data_len)
5140 (_dbus_string_copy_data): new functions
5142 2003-03-28 Anders Carlsson <andersca@codefactory.se>
5144 * dbus/dbus-bus.c: (bus_data_free), (dbus_bus_get):
5148 * dbus/dbus-memory.c:
5151 2003-03-28 Havoc Pennington <hp@pobox.com>
5153 * bus/test.c (bus_test_flush_bus): remove the sleep from here,
5154 I think it may have just been superstition. Not sure.
5156 * dbus/dbus-string.c (_dbus_string_base64_decode): catch some OOM
5157 failures that were not being handled.
5159 * dbus/dbus-auth.c (process_auth): fix a memleak in OOM handling
5161 * dbus/dbus-memory.c: add ability to set number of mallocs in a
5162 row that will fail on out-of-memory.
5164 * dbus/dbus-internals.c (_dbus_test_oom_handling): convenience
5165 function for testing out-of-memory handling.
5167 * bus/config-loader-expat.c (memsuite): don't wrap the dbus
5168 allocation functions, they do map exactly to the expat ones.
5170 2003-03-27 Havoc Pennington <hp@redhat.com>
5172 * bus/config-loader-libxml.c (bus_config_load): add another error
5175 2003-03-26 Anders Carlsson <andersca@codefactory.se>
5178 Add note about automatic service activation.
5180 * doc/dbus-specification.sgml:
5181 Rename the specification and clarify a few things.
5183 2003-03-26 Anders Carlsson <andersca@codefactory.se>
5186 * dbus/dbus-address.c:
5188 * dbus/dbus-marshal.c:
5189 * dbus/dbus-server-debug-pipe.c:
5190 * dbus/dbus-transport-unix.c:
5191 Fix documentation warnings.
5193 2003-03-26 Havoc Pennington <hp@pobox.com>
5195 * bus/test-main.c, dbus/dbus-test.c (main): check memleaks
5196 after every test so it's quick and easy to see which leaked, and
5197 so we test multiple dbus_shutdown() calls
5199 * configure.in: change configure.in XML stuff to also support
5202 * config-loader-libxml.c: some hacking
5204 * config-loader-expat.c: some hacking
5206 * config-parser.c: some hacking, plus tests
5208 2003-03-25 Havoc Pennington <hp@redhat.com>
5210 * throughout - add more _DBUS_ASSERT_ERROR_IS_CLEAR
5212 * configure.in: add --with-xml option to specify XML library,
5213 right now only libxml is supported.
5215 * bus/config-loader-libxml.c, config-parser.c: sync some minor
5216 nonworking code between home and work, still just stubs
5218 2003-03-24 Havoc Pennington <hp@redhat.com>
5220 * dbus/dbus-sysdeps.c (_dbus_set_fd_nonblocking): move to this
5223 * dbus/dbus-errors.c (dbus_set_error, dbus_set_error_const): allow
5224 NULL argument for "message" if the error is a well-known one,
5225 fill in a generic message in this case.
5227 * dbus/dbus-errors.h (DBusResultCode): Kill DBusResultCode in
5230 * bus/test.c (bus_test_flush_bus): add
5232 * bus/policy.c (bus_policy_test): test code stub
5234 2003-03-24 Havoc Pennington <hp@pobox.com>
5236 * bus/connection.c (bus_connections_setup_connection): set up
5237 the "can this user connect" function, but it always returns
5238 TRUE until we have a config file parser so we can have a config
5239 file that allows connections.
5241 2003-03-23 Havoc Pennington <hp@pobox.com>
5243 * dbus/dbus-threads.c (dbus_mutex_new, dbus_condvar_new): with
5244 DBUS_BUILD_TESTS, actually alloc/free a block of memory for
5245 the mutex, so we can check for proper memory management
5248 * dbus/dbus-dataslot.c: remove the mutex from
5249 DBusDataSlotAllocator and lock it manually when using it,
5250 to simplify fitting it into the global slots framework.
5252 * dbus/dbus-threads.c (init_static_locks): rework how we're
5253 handling global locks so they are easily shut down.
5255 * bus/policy.c (bus_policy_append_rule): fix
5257 * bus/test-main.c (main): check for memleaks
5259 * dbus/dbus-test.c (dbus_internal_do_not_use_run_tests): make
5260 test suite check for memleaks
5262 * dbus/dbus-memory.c: add support in test mode for tracking
5263 number of outstanding blocks
5265 2003-03-23 Havoc Pennington <hp@pobox.com>
5267 * bus/policy.c, bus/bus.c, bus/connection.c: implement allow/deny
5270 * dbus/dbus-hash.h: add ULONG hash keys
5272 * dbus/dbus-sysdeps.c (_dbus_get_groups): new
5273 (_dbus_get_group_id): new function
5275 2003-03-20 Havoc Pennington <hp@redhat.com>
5277 * dbus/dbus-connection.c (dbus_connection_set_unix_user_function):
5279 (dbus_connection_get_unix_user): new function
5281 2003-03-20 Havoc Pennington <hp@pobox.com>
5283 * bus/connection.c (bus_connection_send_oom_error): assert that
5284 message has a sender
5285 (connection_execute_transaction): ditto
5286 (bus_connection_preallocate_oom_error): fix to set the sender, and
5287 set recipient to the destination service, not the bus driver
5289 * bus/policy.c: hacking
5291 * dbus/dbus-message.c (dbus_message_service_is): new function
5292 (dbus_message_sender_is): new
5294 2003-03-19 Havoc Pennington <hp@redhat.com>
5296 * bus/policy.c: start sketching code for policy restrictions on
5297 what connections can do.
5299 2003-03-18 Havoc Pennington <hp@redhat.com>
5301 * doc/TODO: some notes on high-level todo items. Little nitpick
5302 stuff is all in @todo, so no need to add it here.
5304 * doc/config-file.txt: some notes on how config file might look
5306 2003-03-18 Anders Carlsson <andersca@codefactory.se>
5312 2003-03-17 Havoc Pennington <hp@redhat.com>
5314 * dbus/dbus-internals.h: add gcc attributes so that
5315 our printf-style functions warn on bad arguments to
5318 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): fix printf
5321 * dbus/dbus-message.c (_dbus_message_loader_queue_messages): fix
5324 2003-03-17 Havoc Pennington <hp@redhat.com>
5326 * bus/test-main.c (main): make it print something as it runs
5327 so make check doesn't look stuck
5329 * doc/negotiation.txt, doc/dbus-sasl-profile.txt: remove
5330 from CVS, now obsolete
5332 2003-03-17 Anders Carlsson <andersca@codefactory.se>
5334 * bus/dispatch.c: (bus_dispatch):
5335 Refetch the service name since it may have been reallocated
5336 when dbus_message_set_sender was called.
5338 * dbus/dbus-sysdeps.c: (_dbus_accept):
5339 Add address and address length variables and use them to stop
5340 valgrind from complaining.
5342 2003-03-17 Havoc Pennington <hp@pobox.com>
5344 All tests pass, no memleaks, no valgrind complaints.
5346 * bus/test.c: refcount handler_slot
5348 * bus/connection.c (bus_connections_new): refcount
5349 connection_data_slot
5351 * dbus/dbus-auth-script.c (_dbus_auth_script_run): delete unused
5352 bytes so that auth scripts pass.
5354 * bus/dispatch.c: init message_handler_slot so it gets allocated
5357 * bus/dispatch.c (message_handler_slot_ref): fix memleak
5359 * dbus/dbus-server-debug-pipe.c (_dbus_server_debug_pipe_new):
5360 dealloc server_pipe_hash when no longer used for benefit of
5363 * dbus/dbus-auth.c (process_command): memleak fix
5365 * bus/dispatch.c (check_hello_message): memleak fix
5367 2003-03-16 Havoc Pennington <hp@pobox.com>
5369 * dbus/dbus-bus.c (ensure_bus_data): fix double-unref of the data slot
5371 2003-03-17 Anders Carlsson <andersca@codefactory.se>
5373 * bus/activation.c (bus_activation_activate_service): Append
5374 the pending activation entry to the list of pending activations.
5376 2003-03-16 Havoc Pennington <hp@pobox.com>
5378 * bus/dispatch.c (bus_dispatch_test): remove double-unrefs of
5381 * dbus/dbus-address.c (create_entry): fix OOM handling when
5382 failing to alloc entry->method
5384 2003-03-16 Havoc Pennington <hp@pobox.com>
5386 * dbus/dbus-watch.c (_dbus_watch_new): handle failure to malloc
5389 * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new):
5390 add some missing dbus_set_result
5392 * bus/dispatch.c (bus_dispatch_add_connection): handle failure to
5393 alloc the DBusMessageHandler
5395 * dbus/dbus-transport.c (_dbus_transport_disconnect): don't ref
5396 the transport here, since we call this from the finalizer; it
5397 resulted in a double-finalize.
5399 * dbus/dbus-transport.c (_dbus_transport_disconnect): fix a bug
5400 where we tried to use transport->connection that was NULL,
5401 happened when transport was disconnected early on due to OOM
5403 * bus/*.c: adapt to handle OOM for watches/timeouts
5405 * dbus/dbus-transport-unix.c: port to handle OOM during
5408 * dbus/dbus-auth.c (_dbus_auth_get_unused_bytes): return a
5409 reference to unused bytes instead of a copy
5411 * dbus/dbus-server.c (dbus_server_handle_watch): return FALSE for
5414 * dbus/dbus-connection.c (dbus_connection_handle_watch): return
5417 * dbus/dbus-timeout.c (dbus_timeout_handle): return FALSE for out
5420 2003-03-16 Anders Carlsson <andersca@codefactory.se>
5422 * doc/dbus-specification.sgml:
5423 Document reply message for ActivateService.
5425 2003-03-16 Anders Carlsson <andersca@codefactory.se>
5427 * bus/activation.c: (bus_pending_activation_entry_free),
5428 (bus_pending_activation_free), (bus_activation_new),
5429 (bus_activation_unref), (bus_activation_service_created),
5430 (bus_activation_activate_service):
5432 * bus/bus.c: (bus_context_new):
5433 * bus/desktop-file.c: (new_section):
5434 * bus/driver.c: (bus_driver_send_service_deleted),
5435 (bus_driver_handle_activate_service):
5436 * bus/services.c: (bus_registry_new), (bus_registry_ensure):
5438 * dbus/dbus-connection.c:
5439 (dbus_connection_send_with_reply_and_block):
5440 * dbus/dbus-message.c: (dbus_message_append_args_valist):
5441 * dbus/dbus-protocol.h:
5442 Make activation work better. Now pending activations will be queued
5443 and the daemon won't try to activate services that are already registered.
5445 2003-03-16 Havoc Pennington <hp@pobox.com>
5447 * dbus/dbus-bus.c (ensure_bus_data): handle failure to set
5450 * dbus/dbus-memory.c (_dbus_initialize_malloc_debug): support
5451 DBUS_MALLOC_BACKTRACES to print trace when failing an alloc
5453 2003-03-16 Havoc Pennington <hp@pobox.com>
5455 * dbus/dbus-string.c (_dbus_string_validate_utf8): oops, unbreak
5456 this. always run the test suite before commit...
5458 * bus/*: adapt to DBusConnection API changes
5460 * glib/dbus-gmain.c: adapt to DBusConnection API changes,
5461 requires renaming stuff to avoid dbus_connection_dispatch name
5464 * dbus/dbus-transport.c (_dbus_transport_queue_messages): new
5467 * dbus/dbus-message.c (_dbus_message_loader_queue_messages):
5468 separate from _dbus_message_loader_return_buffer()
5470 * dbus/dbus-connection.c (dbus_connection_get_n_messages): remove
5471 this, because it's now always broken to use; the number of
5472 messages in queue vs. the number still buffered by the message
5473 loader is undefined/meaningless. Should use
5474 dbus_connection_get_dispatch_state().
5475 (dbus_connection_dispatch): rename from
5476 dbus_connection_dispatch_message
5478 2003-03-16 Havoc Pennington <hp@pobox.com>
5480 * dbus/dbus-string.c (_dbus_string_validate_utf8): copy in a real
5483 2003-03-16 Anders Carlsson <andersca@codefactory.se>
5485 * dbus/dbus-connection.c:
5486 (dbus_connection_send_with_reply_and_block):
5487 Decrease connection->n_incoming when removing an entry
5489 * dbus/dbus-dict.c: (dbus_dict_entry_free),
5490 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
5491 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
5492 (dbus_dict_set_byte_array), (dbus_dict_set_string_array),
5493 (dbus_dict_get_boolean_array), (dbus_dict_get_double_array),
5494 (dbus_dict_get_byte_array):
5495 Handle NULL arrays and strings. Also add support for byte arrays.
5497 * dbus/dbus-marshal.c: (_dbus_marshal_byte_array),
5498 (_dbus_marshal_dict), (_dbus_demarshal_byte_array),
5499 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
5500 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
5501 (_dbus_demarshal_dict), (demarshal_and_validate_len),
5502 (_dbus_marshal_validate_arg), (_dbus_marshal_test):
5503 * dbus/dbus-marshal.h:
5504 Add support for marshalling and demarshalling empty arrays and strings.
5506 * dbus/dbus-message.c: (dbus_message_append_args_valist),
5507 (dbus_message_append_string_array),
5508 (dbus_message_iter_get_boolean),
5509 (dbus_message_iter_get_boolean_array),
5510 (dbus_message_iter_get_int32_array),
5511 (dbus_message_iter_get_uint32_array),
5512 (dbus_message_iter_get_double_array),
5513 (dbus_message_iter_get_byte_array),
5514 (dbus_message_iter_get_string_array), (dbus_message_iter_get_dict),
5515 (check_message_handling):
5516 Add support for getting empty arrays and dicts.
5518 * dbus/dbus-string.c: (_dbus_string_validate_utf8):
5519 Don't do any validation at all for now, that's better than just checking
5522 * test/data/valid-messages/emptiness.message:
5523 New test message with lots of empty arrays.
5525 2003-03-16 Havoc Pennington <hp@pobox.com>
5527 * dbus/dbus-connection.c
5528 (_dbus_connection_queue_received_message_link): new function that
5529 can't fail due to OOM
5531 * dbus/dbus-message.c (_dbus_message_loader_pop_message_link):
5532 new function pops a message together with a list link
5535 * dbus/dbus-transport-unix.c (queue_messages): use new link-based
5536 message queuing functions to avoid needing to alloc memory
5538 2003-03-16 Havoc Pennington <hp@pobox.com>
5540 Oops - test code was only testing failure of around 30 of the
5541 mallocs in the test path, but it turns out there are 500+
5542 mallocs. I believe this was due to misguided linking setup such
5543 that there was one copy of dbus_malloc etc. in the daemon and one
5544 in the shared lib, and only daemon mallocs were tested. In any
5545 case, the test case now tests all 500+ mallocs, and doesn't pass
5546 yet, though there are lots of fixes in this patch.
5548 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
5549 this so that it doesn't need to allocate memory, since it
5550 has no way of indicating failure due to OOM (and would be
5551 annoying if it did).
5553 * dbus/dbus-list.c (_dbus_list_pop_first_link): new function
5555 * bus/Makefile.am: rearrange to create two self-contained
5556 libraries, to avoid having libraries with overlapping symbols.
5557 that was resulting in weirdness, e.g. I'm pretty sure there
5558 were two copies of global static variables.
5560 * dbus/dbus-internals.c: move the malloc debug stuff to
5563 * dbus/dbus-list.c (free_link): free list mempool if it becomes
5566 * dbus/dbus-memory.c (_dbus_disable_mem_pools): new function
5568 * dbus/dbus-address.c (dbus_parse_address): free list nodes
5571 * bus/dispatch.c (bus_dispatch_add_connection): free
5572 message_handler_slot when no longer using it, so
5573 memory leak checkers are happy for the test suite.
5575 * dbus/dbus-server-debug-pipe.c (debug_finalize): free server name
5577 * bus/bus.c (new_connection_callback): disconnect in here if
5578 bus_connections_setup_connection fails.
5580 * bus/connection.c (bus_connections_unref): fix to free the
5582 (bus_connections_setup_connection): if this fails, don't
5583 disconnect the connection, just be sure there are no side
5586 * dbus/dbus-string.c (undo_alignment): unbreak this
5588 * dbus/dbus-auth.c (_dbus_auth_unref): free some stuff we were
5590 (_dbus_auth_new): fix the order in which we free strings
5593 * bus/connection.c (bus_connection_disconnected): fix to
5594 not send ServiceDeleted multiple times in case of memory
5597 * dbus/dbus-bus.c (dbus_bus_get_base_service): new function to
5598 get the base service name
5599 (dbus_bus_register_client): don't return base service name,
5600 instead store it on the DBusConnection and have an accessor
5602 (dbus_bus_register_client): rename dbus_bus_register()
5604 * bus/dispatch.c (check_hello_message): verify that other
5605 connections on the bus also got the correct results, not
5606 just the one sending hello
5608 2003-03-15 Havoc Pennington <hp@pobox.com>
5610 Make it pass the Hello handling test including all OOM codepaths.
5611 Now to do other messages...
5613 * bus/services.c (bus_service_remove_owner): fix crash when
5614 removing owner from an empty list of owners
5615 (bus_registry_ensure): don't leave service in the list of
5616 a connection's owned services if we fail to put the service
5619 * bus/connection.c (bus_connection_preallocate_oom_error): set
5620 error flag on the OOM error.
5622 * dbus/dbus-connection.c (_dbus_connection_new_for_transport):
5623 handle _dbus_transport_set_connection failure
5625 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd): modify
5626 to create watches up front and simply enable/disable them as
5628 (unix_connection_set): this can now fail on OOM
5630 * dbus/dbus-timeout.c, dbus/dbus-watch.c: add concept
5631 of enabling/disabling a watch or timeout.
5633 * bus/loop.c (bus_loop_iterate): don't touch disabled
5636 * glib/dbus-gmain.c: adapt to enable/disable watches and timeouts
5638 2003-03-15 Havoc Pennington <hp@pobox.com>
5640 * bus/dispatch.c (bus_dispatch_test): OK, now finally actually
5641 write useful test code, after all that futzing around ;-)
5643 Test does not yet pass because we can't handle OOM in
5644 _dbus_transport_messages_pending (basically,
5645 dbus_connection_preallocate_send() does not prealloc the write
5646 watch). To fix this, I think we need to add new stuff to
5647 set_watch_functions, namely a SetEnabled function so we can alloc
5648 the watch earlier, then enable it later.
5650 * dbus/Makefile.am (libdbus_convenience_la_SOURCES): move
5651 dbus-memory.c to the convenience lib
5653 * bus/test.c: rename some static functions to keep them clearly
5654 distinct from stuff in connection.c. Handle client disconnection.
5656 2003-03-14 Havoc Pennington <hp@pobox.com>
5658 * bus/dispatch.c (bus_dispatch_test): do test using debug-pipe
5659 transport, tests more of the real codepath. Set up clients
5660 with bus_setup_debug_client.
5662 * bus/test.c (bus_setup_debug_client): function to set up debug
5663 "clients" on the main loop
5665 * dbus/dbus-transport.c (_dbus_transport_open): add debug-pipe
5668 * dbus/dbus-server.c (dbus_server_listen): add debug-pipe
5671 * dbus/dbus-server-debug.c: support a debug server based on pipes
5673 * dbus/dbus-sysdeps.c (_dbus_full_duplex_pipe): new function
5674 (_dbus_close): new function
5676 * configure.in: check for socketpair
5678 2003-03-14 Havoc Pennington <hp@redhat.com>
5680 * dbus/dbus-memory.c: add a "detect buffer overwrites on free"
5683 * dbus/dbus-transport-debug.c: rework this a good bit to be
5684 less complicated. hopefully still works.
5686 * dbus/dbus-server-debug.c (handle_new_client): remove timeout
5689 * glib/dbus-gmain.c (timeout_handler): don't remove timeout
5692 * dbus/dbus-message.c (dbus_message_copy): rename from
5693 dbus_message_new_from_message, fix it up to copy
5694 all the message fields, add test case
5696 * bus/dispatch.c (bus_dispatch_test): add some more test code,
5697 not quite passing yet
5699 2003-03-14 Havoc Pennington <hp@pobox.com>
5701 * bus/loop.c (bus_loop_iterate): add this so we can "run loop
5702 until no work remains" in test code. (the large diff here
5703 is just code movement, no actual changes)
5705 * dbus/dbus-server-debug.c (DEFAULT_INTERVAL): change interval to
5706 1, no point waiting around for test code.
5707 (_dbus_server_debug_accept_transport): unref the timeout
5708 after adding it (right?)
5710 * dbus/dbus-transport-debug.c (DEFAULT_INTERVAL): ditto
5712 2003-03-13 Havoc Pennington <hp@redhat.com>
5714 * dbus/dbus-timeout.c (_dbus_timeout_list_set_functions): handle
5717 * dbus/dbus-watch.c (_dbus_watch_list_set_functions): handle out
5720 * dbus/dbus-connection.h: Make AddWatchFunction and
5721 AddTimeoutFunction return a bool so they can fail on out-of-memory
5723 * bus/bus.c (bus_context_new): set up timeout handlers
5725 * bus/connection.c (bus_connections_setup_connection): set up
5728 * glib/dbus-gmain.c: adapt to the fact that set_functions stuff
5731 * bus/bus.c (bus_context_new): adapt to changes
5733 * bus/connection.c: adapt to changes
5735 * test/watch.c: adapt to DBusWatch changes
5737 * bus/dispatch.c (bus_dispatch_test): started adding this but
5740 2003-03-14 Anders Carlsson <andersca@codefactory.se>
5742 * bus/dispatch.c (send_service_nonexistent_error): Fix typo.
5744 2003-03-13 Havoc Pennington <hp@pobox.com>
5746 * bus/test.c, bus/test.h, bus/Makefile.am, bus/test-main.c:
5747 set up a test framework as for the library
5749 2003-03-12 Havoc Pennington <hp@pobox.com>
5751 Throughout: purge global variables, introduce BusActivation,
5752 BusConnections, BusRegistry, etc. objects instead.
5754 * bus/bus.h, bus/bus.c: introduce BusContext as a global
5757 * test/Makefile.am (TEST_BINARIES): disable bus-test for now,
5758 going to redo this a bit differently I think
5760 2003-03-12 Havoc Pennington <hp@redhat.com>
5762 Mega-patch that gets the message bus daemon initially handling
5763 out-of-memory. Work still needed. Also lots of random
5764 moving stuff to DBusError instead of ResultCode.
5766 * dbus/dbus-list.c (_dbus_list_length_is_one): new function
5768 * dbus/dbus-connection.c
5769 (dbus_connection_send_with_reply_and_block): use DBusError
5771 * dbus/dbus-bus.c: adapt to API changes, make it use DBusError not
5774 * dbus/dbus-connection.c (dbus_connection_send): drop the result
5775 code here, as the only failure possible is OOM.
5777 * bus/connection.c (bus_connection_disconnect):
5778 rename bus_connection_disconnected as it's a notification only
5780 * bus/driver.c (bus_driver_handle_acquire_service): don't free
5781 "name" on get_args failure, should be done by get_args;
5782 don't disconnect client for bad args, just return an error.
5783 (bus_driver_handle_service_exists): ditto
5785 * bus/services.c (bus_services_list): NULL-terminate returned array
5787 * bus/driver.c (bus_driver_send_service_lost)
5788 (bus_driver_send_service_acquired): send messages from driver to a
5789 specific client to the client's unique name, not to the broadcast
5792 * dbus/dbus-message.c (decode_header_data): reject messages that
5793 contain no name field
5794 (_dbus_message_get_client_serial): rename to
5795 dbus_message_get_serial and make public
5796 (_dbus_message_set_serial): rename from set_client_serial
5797 (_dbus_message_set_reply_serial): make public
5798 (_dbus_message_get_reply_serial): make public
5800 * bus/connection.c (bus_connection_foreach): allow stopping
5801 iteration by returning FALSE from foreach function.
5803 * dbus/dbus-connection.c (dbus_connection_send_preallocated)
5804 (dbus_connection_free_preallocated_send)
5805 (dbus_connection_preallocate_send): new API for sending a message
5806 without possibility of malloc failure.
5807 (dbus_connection_send_message): rename to just
5808 dbus_connection_send (and same for whole function family)
5810 * dbus/dbus-errors.c (dbus_error_free): make this reinit the error
5812 * dbus/dbus-sysdeps.c (_dbus_exit): new function
5814 * bus/activation.c: handle/return errors
5816 * dbus/dbus-errors.h: add more DBUS_ERROR #define
5818 * dbus/dbus-sysdeps.c (_dbus_directory_open) (_dbus_file_get_contents)
5819 (_dbus_directory_get_next_file): use DBusError instead of DBusResultCode
5820 (_dbus_result_from_errno): move to this file
5822 2003-03-10 Anders Carlsson <andersca@codefactory.se>
5824 * dbus/dbus-marshal.c:
5825 (_dbus_marshal_set_string):
5826 Take a length argument so we can marshal the correct string
5829 (_dbus_marshal_dict), (_dbus_demarshal_dict),
5830 (_dbus_marshal_get_arg_end_pos), (_dbus_marshal_validate_arg),
5831 (_dbus_marshal_test):
5832 * dbus/dbus-marshal.h:
5833 Add support for marshalling and demarshalling dicts.
5835 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
5836 Add support for TYPE DICT.
5838 * dbus/dbus-message.c: (set_string_field):
5839 Adjust header padding.
5841 (dbus_message_append_args_valist), (dbus_message_append_dict),
5842 (dbus_message_get_args_valist), (dbus_message_iter_get_arg_type),
5843 (dbus_message_iter_get_dict), (_dbus_message_loader_return_buffer),
5844 (check_message_handling), (check_have_valid_message):
5845 * dbus/dbus-message.h:
5846 Add functions for setting and getting dicts.
5848 * dbus/dbus-protocol.h:
5854 * doc/dbus-specification.sgml:
5855 Add information about how dicts are marshalled.
5857 * test/data/invalid-messages/dict-with-nil-value.message:
5858 * test/data/invalid-messages/too-short-dict.message:
5859 * test/data/valid-messages/dict-simple.message:
5860 * test/data/valid-messages/dict.message:
5861 Add sample messages containing dicts.
5863 2003-03-08 Anders Carlsson <andersca@codefactory.se>
5865 * dbus/dbus-dict.h: Add DBUS_END_DECLS.
5867 2003-03-07 Anders Carlsson <andersca@codefactory.se>
5870 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
5871 (dbus_dict_get_keys), (insert_entry), (dbus_dict_set_boolean),
5872 (dbus_dict_set_int32), (dbus_dict_set_uint32),
5873 (dbus_dict_set_double), (dbus_dict_set_string),
5874 (dbus_dict_set_boolean_array), (dbus_dict_set_int32_array),
5875 (dbus_dict_set_uint32_array), (dbus_dict_set_double_array),
5876 (dbus_dict_set_string_array), (_dbus_dict_test):
5878 Fix according to comments from Havoc.
5880 2003-03-06 Michael Meeks <michael@server.home>
5882 * configure.in: if we don't have kde-config, disable have_qt.
5884 2003-03-07 Anders Carlsson <andersca@codefactory.se>
5889 * dbus/dbus-dict.c: (dbus_dict_entry_free), (dbus_dict_new),
5890 (dbus_dict_ref), (dbus_dict_unref), (dbus_dict_contains),
5891 (dbus_dict_remove), (dbus_dict_get_value_type),
5892 (dbus_dict_get_keys), (dbus_dict_put_boolean),
5893 (dbus_dict_put_int32), (dbus_dict_put_uint32),
5894 (dbus_dict_put_double), (dbus_dict_put_string),
5895 (dbus_dict_put_boolean_array), (dbus_dict_put_int32_array),
5896 (dbus_dict_put_uint32_array), (dbus_dict_put_double_array),
5897 (dbus_dict_put_string_array), (dbus_dict_get_boolean),
5898 (dbus_dict_get_int32), (dbus_dict_get_uint32),
5899 (dbus_dict_get_double), (dbus_dict_get_string),
5900 (dbus_dict_get_boolean_array), (dbus_dict_get_int32_array),
5901 (dbus_dict_get_uint32_array), (dbus_dict_get_double_array),
5902 (dbus_dict_get_string_array), (_dbus_dict_test):
5904 Add DBusDict implementation
5906 * dbus/dbus-test.c: (dbus_internal_do_not_use_run_tests):
5910 2003-03-04 Havoc Pennington <hp@pobox.com>
5912 * test/data/auth/*: adapt to changes
5914 * dbus/dbus-auth-script.c (_dbus_auth_script_run): add
5915 USERID_BASE64 and change USERNAME_BASE64 to put in username not
5918 * dbus/dbus-keyring.c (_dbus_keyring_validate_context): prevent
5919 more stuff from being in a context name, to make the protocol
5920 simpler to deal with
5922 * dbus/dbus-errors.c (dbus_error_has_name): new function
5923 (dbus_error_is_set): new function
5925 * dbus/dbus-auth.c: replace DBUS_STUPID_TEST_MECH auth
5926 with DBUS_COOKIE_SHA1, implement DBUS_COOKIE_SHA1
5928 * dbus/dbus-connection.c (dbus_connection_flush): also read
5929 messages during a flush operation
5931 * dbus/Makefile.am: remove dbus-md5 since it isn't currently used.
5933 2003-03-05 Anders Carlsson <andersca@codefactory.se>
5935 * configure.in: Check for gethostbyname on Solaris.
5937 * dbus/dbus-transport.c: (_dbus_transport_open):
5938 Remove duplicate "tcp" entry.
5940 * doc/dbus-specification.sgml:
5941 Clarify some things.
5943 2003-03-05 Anders Carlsson <andersca@codefactory.se>
5945 * dbus/dbus-auth.c: (send_rejected), (process_test_subdir):
5946 * dbus/dbus-keyring.c: (_dbus_keyring_new_homedir),
5947 (_dbus_keyring_test):
5948 * dbus/dbus-md5.c: (_dbus_md5_compute):
5949 * dbus/dbus-sha.c: (_dbus_sha_compute):
5952 2003-03-05 Anders Carlsson <andersca@codefactory.se>
5954 * README: Add some things.
5956 2003-03-04 Anders Carlsson <andersca@codefactory.se>
5958 * dbus/dbus-message.c (dbus_message_append_args_valist): Add a break;
5959 after case DBUS_TYPE_BOOELAN.
5961 2003-03-02 Havoc Pennington <hp@pobox.com>
5963 * test/break-loader.c (randomly_set_extreme_ints): add test that
5964 sets really huge and small integers
5966 * dbus/dbus-marshal.c (_dbus_marshal_validate_arg): add check
5967 that length of boolean array fits in the string, and that
5968 string has room for boolean value in single-bool case.
5970 * dbus/dbus-message-builder.c (_dbus_message_data_load): add
5971 optional value to "ALIGN" command which is what to fill the
5974 * test/data/valid-messages/no-padding.message: add regression
5975 test for the message padding problem
5977 2003-03-02 Havoc Pennington <hp@pobox.com>
5979 * dbus/dbus-message.c (decode_header_data): fix to always init
5980 message_padding, from Benjamin Dauvergne
5982 2003-03-02 Havoc Pennington <hp@pobox.com>
5988 2003-03-01 Joe Shaw <joe@assbarn.com>
5990 * configure.in: Check for "struct cmsgcred" and try to access its
5991 members for BSD-like unices.
5993 * dbus/dbus-sysdeps.c (read_credentials_byte): Fold this back into
5994 _dbus_read_credentials_unix_socket().
5995 (_dbus_read_credentials_unix_socket): Use recvmsg() instead of
5996 read() for reading the credential byte off the unix socket. Use
5997 struct cmsgcred on systems that support it.
5999 2003-02-27 Alexander Larsson <alexl@redhat.com>
6003 Make gthreads-2.0 dependency optional. Don't build thread test if
6006 2003-02-27 Havoc Pennington <hp@pobox.com>
6008 * dbus/dbus-connection.c
6009 (dbus_connection_send_message_with_reply_and_block): fix doh!
6010 doh! doh! bug that resulted in never removing a reply from the
6011 queue, no wonder we called get_reply_serial so much ;-)
6013 * dbus/dbus-message.c (struct DBusMessage): cache reply serial
6014 and client serial instead of demarshaling them every time
6016 2003-02-27 Havoc Pennington <hp@pobox.com>
6018 * dbus/dbus-marshal.c (_dbus_demarshal_int32): rewrite to be much
6019 more inlined, using dbus-string-private.h, speeds things up
6022 * dbus/dbus-string.c (_dbus_string_free): apply align offset
6023 when freeing the string
6024 (_dbus_string_steal_data): fix for align offset
6025 (undo_alignment): new function
6027 2003-02-26 Havoc Pennington <hp@redhat.com>
6029 All kinds of audit fixes from Owen, plus initial attempt to
6030 handle unaligned memory returned from malloc.
6032 * dbus/dbus-string.c (_dbus_string_init): clamp max length to
6033 leave room for align_offset and nul byte
6034 (fixup_alignment): function to track an align_offset and
6035 ensure real->str is aligned
6036 (DBUS_GENERIC_STRING_PREAMBLE): len must be less than allocated,
6037 to allow a nul byte plus align offset
6038 (_dbus_string_lock): fix overflow issue
6039 (_dbus_string_init_const_len): add assertions on sanity of len,
6040 assign allocated to be ALLOCATION_PADDING larger than len
6041 (set_length): fixup the overflow handling
6042 (_dbus_string_get_data_len): fix overflow in assertion
6043 (open_gap): detect overflow in size of gap to be opened
6044 (_dbus_string_lengthen): add overflow check
6045 (_dbus_string_align_length): fix overflow with _DBUS_ALIGN_VALUE
6046 (_dbus_string_append): add overflow check
6047 (_dbus_string_append_unichar): overflow
6048 (_dbus_string_delete): fix overflow in assertion
6049 (_dbus_string_copy_len): overflow in assertion
6050 (_dbus_string_replace_len): overflows in assertions
6051 (_dbus_string_find): change to implement in terms of
6052 _dbus_string_find_to
6053 (_dbus_string_find_to): assorted fixage
6054 (_dbus_string_equal_c_str): assert c_str != NULL,
6055 fix logic so the function works
6056 (_dbus_string_ends_with_c_str): fix overflow thingy
6057 (_dbus_string_base64_encode): overflow fix
6058 (_dbus_string_validate_ascii): overflow
6059 (_dbus_string_validate_nul): overflow
6061 2003-02-26 Havoc Pennington <hp@redhat.com>
6063 * dbus/dbus-marshal.c (_dbus_marshal_test): fix to work with DISABLE_ASSERTS
6065 2003-02-26 Alexander Larsson <alexl@redhat.com>
6068 Set DBUS_GLIB_THREADS_LIBS for apps using gthread-2.0
6070 * dbus/dbus-connection.c:
6071 * dbus/dbus-connection.h:
6072 Fix _dbus_connection_acquire_io_path and _dbus_connection_acquire_dispatch.
6073 Add dbus_connection_set_wakeup_main_function and use it when queueing
6074 incoming and outgoing messages.
6077 * dbus/dbus-dataslot.c:
6078 Threadsafe usage of DBusDataSlotAllocator
6080 * dbus/dbus-message.c: (dbus_message_get_args_iter):
6083 * dbus/dbus-server-unix.c:
6086 * glib/dbus-gmain.c:
6087 Implement the new wakeup functions for glib.
6090 * glib/test-thread-client.c:
6091 * glib/test-thread-server.c:
6092 * glib/test-thread.h:
6093 Initial cut at some thread test code. Not really done yet.
6095 2003-02-26 Havoc Pennington <hp@pobox.com>
6097 * dbus/dbus-connection.c
6098 (dbus_connection_send_message_with_reply_and_block): fix crash
6099 where we ref'd the outgoing message instead of the returned reply
6101 * dbus/dbus-transport-unix.c (do_authentication): check read watch
6102 at the end of this function, so if we didn't need to read for
6103 authentication, we reinstall it for receiving messages
6105 * dbus/dbus-message.c (dbus_message_new_reply): allow replies to
6106 a NULL sender for peer-to-peer case
6108 * dbus/dbus-transport-unix.c (check_read_watch): handle
6109 !authenticated case correctly
6111 * glib/dbus-gmain.c: add support for DBusServer
6113 * dbus/dbus-server.c: add data slot support
6115 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): check
6116 return values and handle errors
6118 * dbus/dbus-dataslot.c: factor out the data slot stuff from
6121 * Doxyfile.in (INPUT): add glib subdir
6123 * glib/dbus-gmain.c (dbus_connection_setup_with_g_main): rename
6124 setup_with_g_main instead of hookup_with_g_main; write docs
6126 2003-02-24 Anders Carlsson <andersca@codefactory.se>
6128 * dbus/dbus-marshal.c: (_dbus_marshal_validate_arg):
6129 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
6130 * dbus/dbus-message.c: (dbus_message_append_boolean),
6131 (dbus_message_append_boolean_array),
6132 (dbus_message_get_args_valist), (_dbus_message_test):
6133 * dbus/dbus-message.h:
6134 * doc/dbus-specification.sgml:
6135 Various fixes as pointed out by Havoc.
6137 * test/data/invalid-messages/bad-boolean-array.message:
6138 * test/data/invalid-messages/bad-boolean.message:
6139 Add invalid boolean value test cases.
6141 2003-02-24 Anders Carlsson <andersca@codefactory.se>
6143 * dbus/dbus-internals.c: (_dbus_type_to_string):
6144 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
6145 (_dbus_marshal_validate_arg):
6146 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
6147 * dbus/dbus-message.c: (dbus_message_append_args_valist),
6148 (dbus_message_append_boolean), (dbus_message_append_boolean_array),
6149 (dbus_message_get_args_valist), (dbus_message_iter_get_boolean),
6150 (dbus_message_iter_get_int32), (dbus_message_iter_get_uint32),
6151 (dbus_message_iter_get_double),
6152 (dbus_message_iter_get_boolean_array), (message_iter_test):
6153 * dbus/dbus-message.h:
6154 * dbus/dbus-protocol.h:
6155 * doc/dbus-specification.sgml:
6156 * test/data/valid-messages/lots-of-arguments.message:
6157 Add support for boolean and boolean array types.
6159 2003-02-23 Havoc Pennington <hp@pobox.com>
6161 * dbus/dbus-keyring.c: finish most of this implementation and
6164 * dbus/dbus-errors.c (dbus_set_error_const, dbus_set_error): make
6165 these barf if the error isn't cleared to NULL
6167 * dbus/dbus-sysdeps.c (_dbus_delete_file): set error on failure
6168 (_dbus_create_directory): new function
6170 * dbus/dbus-errors.c (dbus_set_error): fix warning
6172 * dbus/dbus-string.c (_dbus_string_hex_encode): new function
6173 (_dbus_string_hex_decode): new function
6174 (test_hex_roundtrip): test code
6176 * dbus/dbus-sha.c (_dbus_sha_compute): use dbus_string_hex_encode
6178 * dbus/dbus-md5.c (_dbus_md5_compute): use dbus_string_hex_encode
6180 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): make this use
6181 the save-to-temp/rename trick to atomically write the new file
6182 (_dbus_string_parse_uint): new function
6184 2003-02-22 Havoc Pennington <hp@pobox.com>
6186 * test/Makefile.am (dist-hook): fix dist for test/data/sha-1
6188 2003-02-22 Havoc Pennington <hp@pobox.com>
6190 * dbus/dbus-message.c (dbus_message_iter_get_string_array):
6191 (dbus_message_iter_get_byte_array): Fix up doxygen warnings
6193 * dbus/dbus-sha.c: add implementation of SHA-1 algorithm
6195 * dbus/test/data/sha-1: add US government test suite for SHA-1
6197 2003-02-21 Anders Carlsson <andersca@codefactory.se>
6199 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
6200 Make string arrays NULL-terminated.
6202 * dbus/dbus-memory.c: (dbus_free_string_array):
6203 * dbus/dbus-memory.h:
6204 New function for freeing NULL-terminated string arrays.
6206 * dbus/dbus-message-builder.c: (append_quoted_string),
6207 (_dbus_message_data_load):
6208 Add support for array types.
6210 * dbus/dbus-message.c: (check_message_handling):
6211 Add more types as test cases.
6213 * dbus/dbus-sysdeps.c: (_dbus_string_parse_int),
6214 (_dbus_string_parse_double):
6215 Add the start offset to the end offset.
6217 * test/data/valid-messages/lots-of-arguments.message:
6218 New test message with lots of arguments.
6220 2003-02-21 Anders Carlsson <andersca@codefactory.se>
6222 * dbus/dbus-message.c: (dbus_message_append_nil),
6223 (dbus_message_append_int32), (dbus_message_append_uint32),
6224 (dbus_message_append_double), (dbus_message_append_string),
6225 (dbus_message_append_int32_array),
6226 (dbus_message_append_uint32_array),
6227 (dbus_message_append_double_array),
6228 (dbus_message_append_byte_array),
6229 (dbus_message_append_string_array):
6230 Fix all out-of-memory handling in these functions.
6232 2003-02-21 Anders Carlsson <andersca@codefactory.se>
6234 * dbus/dbus-message.c: (dbus_message_append_nil):
6237 2003-02-21 Anders Carlsson <andersca@codefactory.se>
6239 * dbus/dbus-message.c: (dbus_message_append_args_valist),
6240 (dbus_message_append_nil), (dbus_message_append_int32_array),
6241 (dbus_message_append_uint32_array),
6242 (dbus_message_append_double_array),
6243 (dbus_message_append_byte_array),
6244 (dbus_message_append_string_array), (dbus_message_get_args_valist),
6245 (dbus_message_iter_get_int32_array),
6246 (dbus_message_iter_get_uint32_array),
6247 (dbus_message_iter_get_double_array),
6248 (dbus_message_iter_get_byte_array),
6249 (dbus_message_iter_get_string_array):
6251 * dbus/dbus-message.h:
6252 Add functions for appending and getting arrays.
6254 2003-02-21 Anders Carlsson <andersca@codefactory.se>
6256 * dbus/dbus-mempool.c (_dbus_mem_pool_new): Make the
6257 element size at least 8 bytes, fixes mempool tests on
6260 2003-02-20 Alexander Larsson <alexl@redhat.com>
6262 * dbus/dbus-transport-unix.c (unix_do_iteration):
6263 Unlock the connection mutex during a blocking select call.
6264 Add todo about how we need a way to wake up the select.
6266 * dbus/dbus-connection-internal.h:
6267 * dbus/dbus-connection.c:
6268 Add _dbus_connection_lock and _dbus_connection_unlock.
6270 2003-02-19 Havoc Pennington <hp@pobox.com>
6272 * Doxyfile.in (PREDEFINED): put DOXYGEN_SHOULD_SKIP_THIS in
6273 Doxyfile.in, not Doxyfile
6275 * dbus/dbus-keyring.c: do some hacking on this
6277 * dbus/dbus-sysdeps.c (_dbus_delete_file): new
6279 * dbus/dbus-errors.c (dbus_set_error_const): do not call
6281 (dbus_set_error): remove dbus_error_init, check for message ==
6282 NULL *before* we sprintf into it, and add @todo about including
6283 system headers in this file
6285 * dbus/dbus-sysdeps.c (_dbus_create_file_exclusively): new
6287 * dbus/dbus-errors.h (DBUS_ERROR_FAILED): add
6289 * dbus/dbus-sysdeps.c (get_user_info): break this function out to
6290 get various bits of user information based on either username
6292 (_dbus_homedir_from_username): new function
6294 2003-02-19 Anders Carlsson <andersca@codefactory.se>
6297 Add check for nonposix getpwnam_r
6299 * dbus/dbus-mempool.c: (_dbus_mem_pool_new):
6300 Align the pool element size to a sizeof (void *) boundary.
6302 * dbus/dbus-sysdeps.c: (_dbus_setenv), (_dbus_connect_unix_socket),
6303 (_dbus_listen_unix_socket), (_dbus_credentials_from_username):
6304 General Solaris fixes.
6306 * test/data/valid-messages/simplest-manual.message:
6307 Explicitly state that we want little-endian packing.
6309 2003-02-19 Mikael Hallendal <micke@codefactory.se>
6311 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
6313 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_tcp_socket):
6314 Added to create a transport connecting using a tcp/ip socket.
6316 * dbus/dbus-sysdeps.c (_dbus_connect_tcp_socket): Added to connect
6317 to a tcp socket at given host and port.
6318 (_dbus_listen_tcp_socket): added to listen on tcp socket for given
6321 * dbus/dbus-server.c (dbus_server_listen): Support tcp: addresses.
6323 * dbus/dbus-server-unix.c (_dbus_server_new_for_tcp_socket):
6324 Added to create a server listening on a TCP/IP socket.
6326 2003-02-19 Havoc Pennington <hp@pobox.com>
6328 Throughout: mop up all the Doxygen warnings and undocumented
6331 * dbus/dbus-sysdeps.c (do_exec): do not use execvp, we don't want
6332 to search any paths.
6334 * dbus/dbus-threads.c: move global mutex initializers to
6335 dbus-internals.h, multiple prototypes was confusing doxygen
6336 besides being kind of ugly
6338 * Doxyfile (PREDEFINED): have Doxygen define
6339 DOXYGEN_SHOULD_SKIP_THIS so we can exclude things from
6340 docs with #ifndef DOXYGEN_SHOULD_SKIP_THIS
6341 (do not abuse the feature! it's for stuff like the autogenerated
6342 macros in dbus-md5.c, not just for things you don't feel like
6345 2003-02-18 Havoc Pennington <hp@pobox.com>
6347 * dbus/dbus-string.c (_dbus_string_zero): new function
6349 * dbus/dbus-md5.c: include MD5 implementation by L. Peter Deutsch,
6350 wrap it in some dbus-friendly API
6352 * dbus/dbus-types.h: add 16-bit types
6354 2003-02-18 Joe Shaw <joe@assbarn.com>
6356 * dbus/dbus-auth.c (handle_server_data_stupid_test_mech): Just get
6357 credentials from our currently running process.
6358 (get_word): Fix a buglet where we were copying the entire length
6359 instead of relative to our position.
6361 * dbus/dbus-hash.c (_dbus_hash_test): Don't try to allocate the
6362 keys on the stack... it's 640k of data.
6364 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): Always
6365 read the credentials byte off the socket, even if we don't have
6367 (_dbus_poll): Implement poll() using select() for systems which
6370 * glib/test-dbus-glib.c (main): Print out an error if no
6371 parameters are given.
6373 * test/data/auth/fallback.auth-script: Added. Tests that a client
6374 can fallback to a secondary auth mechanism if the first fails.
6376 2003-02-18 Havoc Pennington <hp@pobox.com>
6380 2003-02-17 Havoc Pennington <hp@pobox.com>
6382 * doc/dbus-specification.sgml: lots of cosmetic
6383 cleanups/rearrangement, add assorted FIXME, change DBUS_ADDRESS
6384 env variable to DBUS_BUS_ADDRESS, s/client/application/,
6385 s/server/bus/ (except in authentication section). Add a section
6386 "Message Bus Message Routing"
6388 2003-02-17 Anders Carlsson <andersca@codefactory.se.>
6394 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6396 * doc/dbus-specification.sgml:
6397 Specification updates.
6399 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6401 * bus/activation.c: (bus_activation_init), (child_setup),
6402 (bus_activation_activate_service):
6404 * bus/main.c: (main):
6405 Set DBUS_ADDRESS environment variable.
6407 * dbus/dbus-errors.c: (dbus_set_error):
6408 Don't use va_copy since that's a C99 feature.
6410 * dbus/dbus-sysdeps.c: (_dbus_setenv), (do_exec),
6411 (_dbus_spawn_async):
6412 * dbus/dbus-sysdeps.h:
6413 Add child_setup_func to _dbus_spawn_async.
6415 * doc/dbus-specification.sgml:
6416 Update specification.
6418 * test/spawn-test.c: (setup_func), (main):
6421 2003-02-17 Alexander Larsson <alexl@redhat.com>
6423 * dbus/dbus-connection.c (_dbus_connection_handler_destroyed_locked):
6426 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6430 * doc/dbus-test-plan.sgml:
6431 Add test plan document.
6436 2003-02-17 Anders Carlsson <andersca@codefactory.se>
6438 * dbus/dbus-message.c: (decode_header_data),
6439 (_dbus_message_loader_return_buffer):
6440 Set the header padding amount when loading a message.
6442 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6444 * bus/dispatch.c: (send_one_message):
6445 Only send broadcast messages to registered connections.
6447 * dbus/dbus-message.c: (dbus_message_name_is):
6448 * dbus/dbus-message.h:
6449 New convenience function.
6451 * dbus/dbus-transport-debug.c: (do_reading):
6452 Only dispatch one message per run.
6455 * test/bus-test.c: (new_connection_callback), (die),
6456 (test_hello_client1_handler), (test_hello_client2_handler),
6457 (test_hello_replies), (main):
6459 * test/bus-test-loop.[ch]:
6462 2003-02-16 Havoc Pennington <hp@pobox.com>
6464 * dbus/dbus-connection.c (dbus_connection_dispatch_message): fix
6465 backward conditional
6467 2003-02-16 Alexander Larsson <alexl@redhat.com>
6469 * dbus/dbus-connection.c:
6470 Implement sent_message_with_reply. (with_reply_and block is still
6472 Made dispatch_message not lose message if OOM.
6474 * dbus/dbus-errors.h:
6475 Add NoReply error (for reply timeouts).
6477 2003-02-16 Alexander Larsson <alexl@redhat.com>
6479 * dbus/dbus-hash.c (_dbus_hash_table_unref):
6480 Actually free keys and values when destroying hashtable.
6482 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6484 * dbus/dbus-auth.c: (client_try_next_mechanism):
6487 * dbus/dbus-threads.c: (dbus_condvar_wait_timeout):
6488 Return TRUE if there's no thread implementation around.
6490 * glib/dbus-gmain.c: (free_source),
6491 (dbus_connection_hookup_with_g_main):
6492 Make sure to remove the GSource when the connection is finalized.
6494 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6496 * bus/dispatch.c: (bus_dispatch_message_handler):
6497 * dbus/dbus-errors.h:
6498 Return an error if someone tries to send a message to a service
6501 2003-02-16 Anders Carlsson <andersca@codefactory.se>
6503 * bus/activation.c: (load_directory), (bus_activation_init),
6504 (bus_activation_activate_service):
6507 (bus_driver_handle_activate_service), (bus_driver_handle_message):
6508 More work on the activation handling.
6510 * dbus/dbus-errors.h:
6511 Add some error messages
6513 * dbus/dbus-message.c: (dbus_message_new_error_reply):
6514 * dbus/dbus-message.h:
6515 New function that creates an error message.
6517 * dbus/dbus-protocol.h:
6518 Add ACTIVATE_SERVER message.
6520 * dbus/dbus-server-unix.c: (unix_handle_watch),
6521 (_dbus_server_new_for_domain_socket):
6522 Call _dbus_fd_set_close_on_exec.
6524 * dbus/dbus-sysdeps.c: (make_pipe), (do_exec),
6525 (_dbus_spawn_async), (_dbus_disable_sigpipe),
6526 (_dbus_fd_set_close_on_exec):
6527 * dbus/dbus-sysdeps.h:
6528 Add _dbus_fd_set_close_on exec function. Also add function that checks
6529 that all open fds are set to close-on-exec and warns otherwise.
6531 * dbus/dbus-transport-unix.c:
6532 (_dbus_transport_new_for_domain_socket):
6533 Call _dbus_fd_set_close_on_exec.
6535 2003-02-16 Havoc Pennington <hp@pobox.com>
6537 * dbus/dbus-connection.c (dbus_connection_set_change_sigpipe):
6538 allow people to avoid setting SIGPIPE to SIG_IGN
6539 (_dbus_connection_new_for_transport): disable SIGPIPE unless
6540 we've been asked not to
6542 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6544 * dbus/dbus-list.c: (_dbus_list_append_link),
6545 (_dbus_list_prepend_link):
6546 * dbus/dbus-memory.c: (dbus_malloc), (dbus_malloc0),
6550 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6553 * bus/activation.c: (bus_activation_entry_free),
6554 (add_desktop_file_entry), (load_directory), (bus_activation_init):
6556 * bus/main.c: (main):
6557 Add simple activation support, doesn't work yet though.
6559 2003-02-15 Zack Rusin <zack@kde.org>
6561 * qt/dbus-qthread.cpp: small casting fix
6563 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6565 * dbus/dbus-errors.c: (dbus_set_error):
6566 * dbus/dbus-errors.h:
6567 Add a few errors and make dbus_set_error void.
6569 * dbus/dbus-sysdeps.c:
6570 (_dbus_errno_to_string), (close_and_invalidate), (make_pipe),
6571 (write_err_and_exit), (read_ints), (do_exec), (_dbus_spawn_async):
6572 * dbus/dbus-sysdeps.h:
6573 Add _dbus_spawn_async.
6575 * test/spawn-test.c: (main):
6576 Test for _dbus_spawn_async.
6578 2003-02-15 Anders Carlsson <andersca@codefactory.se>
6580 * dbus/dbus-internals.h:
6581 Fix build without tests.
6583 * dbus/dbus-list.c: (alloc_link):
6584 Fix a segfault when a malloc fails.
6586 * dbus/dbus-memory.c: (initialize_malloc_debug), (dbus_malloc),
6587 (dbus_malloc0), (dbus_realloc):
6588 Add support for malloc debugging.
6590 2003-02-15 Alexander Larsson <alexl@redhat.com>
6592 * dbus/dbus-threads.c:
6593 * dbus/dbus-threads.h:
6594 Add condvars. Remove static mutext from API.
6595 Implement static mutexes by initializing them from threads_init.
6597 * glib/dbus-gthread.c:
6598 * qt/dbus-qthread.cpp:
6599 Update with the thread api changes.
6604 Turn StaticMutex into normal mutex + init function.
6605 Export new functions _dbus_list_alloc_link, _dbus_list_free_link,
6606 _dbus_list_append_link, _dbus_list_prepend_link
6609 * dbus/dbus-sysdeps.c:
6610 * dbus/dbus-sysdeps.h:
6611 New type dbus_atomic_t, and new functions _dbus_atomic_inc,
6612 _dbus_atomic_dec. Only slow fallback implementation at the moment.
6614 * dbus/dbus-protocol.h:
6615 Add DBUS_MESSAGE_LOCAL_DISCONNECT define
6617 * dbus/dbus-message.c:
6618 Make ref/unref atomic.
6621 * dbus/dbus-connection-internal.h:
6622 * dbus/dbus-connection.c:
6623 * dbus/dbus-connection.h:
6625 Change _peek to _borrow,_return & _steal_borrowed.
6626 Change disconnect callback to event.
6627 Make dbus_connection_dispatch_messages reentrant.
6629 * dbus/dbus-transport.c:
6630 Don't ref the connection on calls to the transport
6633 * dbus/dbus-message-handler.c:
6636 * glib/dbus-gmain.c:
6637 Don't use peek_message anymore
6640 * test/debug-thread.c:
6641 * test/debug-thread.h:
6642 Simple thread implementation that asserts() on deadlocks in
6643 single-threaded code.
6646 (main) Call debug_threads_init.
6649 Use disconnect message instead of disconnect callback.
6653 Don't call dbus_connection_set_disconnect_function. Instead export
6654 bus_connection_disconnect.
6657 Call bus_connection_disconnect when we get a disconnected message.
6659 2003-02-15 Havoc Pennington <hp@pobox.com>
6661 * dbus/dbus-message.c (dbus_message_new): fool around with the
6664 2003-02-14 Havoc Pennington <hp@pobox.com>
6666 * dbus/dbus-mempool.c: fail if the debug functions so indicate
6668 * dbus/dbus-memory.c: fail if the debug functions indicate we
6671 * dbus/dbus-internals.c (_dbus_set_fail_alloc_counter)
6672 (_dbus_decrement_fail_alloc_counter): debug functions to
6673 simulate memory allocation failures
6675 2003-02-14 Havoc Pennington <hp@pobox.com>
6677 * dbus/dbus-errors.h (struct DBusError): add a word of padding
6680 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6682 * bus/driver.c: (bus_driver_handle_hello):
6684 * bus/services.c: (bus_service_lookup):
6685 Reorder message sending so we get a more sane order.
6687 * test/bus-test.c: (message_handler):
6690 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6692 * bus/driver.c: (bus_driver_send_service_deleted),
6693 (bus_driver_send_service_created), (bus_driver_send_service_lost),
6694 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
6695 (bus_driver_send_welcome_message),
6696 (bus_driver_handle_list_services),
6697 (bus_driver_handle_acquire_service),
6698 (bus_driver_handle_service_exists):
6699 * dbus/dbus-bus.c: (dbus_bus_register_client),
6700 (dbus_bus_acquire_service), (dbus_bus_service_exists):
6701 * dbus/dbus-errors.c: (dbus_result_to_string):
6702 * dbus/dbus-errors.h:
6703 * dbus/dbus-message.c: (dbus_message_append_args),
6704 (dbus_message_append_args_valist), (dbus_message_get_args),
6705 (dbus_message_get_args_valist), (dbus_message_get_args_iter),
6706 (dbus_message_iter_get_arg_type), (dbus_message_iter_get_string),
6707 (dbus_message_iter_get_byte_array),
6708 (dbus_message_iter_get_string_array), (message_iter_test),
6709 (check_message_handling), (_dbus_message_test):
6710 * dbus/dbus-message.h:
6711 * test/bus-test.c: (main):
6712 Change fields to arguments in messages, so that they won't be
6713 confused with header fields.
6715 * glib/test-dbus-glib.c: (main):
6716 Remove append_fields from hello message.
6718 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6720 * dbus/dbus-errors.c:
6721 * dbus/dbus-message.c:
6722 * dbus/dbus-string.c:
6723 Documentation fixes.
6725 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6727 * glib/dbus-gmain.c: (timeout_handler), (add_timeout),
6729 Implement support for timeouts in dbus-glib.
6731 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6733 * dbus/dbus-message-builder.c: (_dbus_message_data_load):
6734 * dbus/dbus-message.c: (process_test_subdir):
6735 * test/break-loader.c: (find_breaks_based_on):
6736 Plug some memory leaks.
6738 2003-02-13 Richard Hult <rhult@codefactory.se>
6740 * bus/main.c: Fix build.
6742 * dbus/dbus-errors.h:
6743 * dbus/dbus-errors.c: Fix copyright for Anders.
6745 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6750 * bus/connection.c: (bus_connection_foreach):
6753 * bus/desktop-file.c: (grow_lines_in_section), (grow_sections),
6754 (unescape_string), (new_section), (parse_section_start),
6755 (parse_key_value), (report_error), (bus_desktop_file_load),
6756 (bus_desktop_file_get_string):
6757 * bus/desktop-file.h:
6758 Use DBusError for error reporting.
6760 * bus/dispatch.c: (send_one_message),
6761 (bus_dispatch_message_handler):
6762 * bus/driver.c: (bus_driver_send_service_deleted),
6763 (bus_driver_send_service_created), (bus_driver_send_service_lost),
6764 (bus_driver_send_service_acquired), (bus_driver_handle_hello),
6765 (bus_driver_send_welcome_message),
6766 (bus_driver_handle_list_services),
6767 (bus_driver_handle_acquire_service),
6768 (bus_driver_handle_service_exists):
6769 * bus/loop.c: (bus_loop_run):
6771 Use BUS_HANDLE_OOM instead of _DBUS_HANDLE_OOM.
6773 * bus/utils.c: (bus_wait_for_memory):
6775 New files with general utility functions.
6777 * dbus/dbus-internals.h:
6778 Remove _DBUS_HANDLE_OOM.
6780 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6782 * dbus/dbus-errors.c: (dbus_result_to_string), (dbus_error_init),
6783 (dbus_error_free), (dbus_set_error_const), (dbus_set_error):
6784 * dbus/dbus-errors.h:
6785 Add DBusError structure.
6787 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6789 * test/data/valid-messages/standard-acquire-service.message:
6790 * test/data/valid-messages/standard-hello.message:
6791 * test/data/valid-messages/standard-list-services.message:
6792 * test/data/valid-messages/standard-service-exists.message:
6793 Add some standard messages.
6795 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6797 * bus/driver.c: (bus_driver_send_welcome_message),
6798 (bus_driver_handle_list_services),
6799 (bus_driver_handle_acquire_service),
6800 (bus_driver_handle_service_exists), (bus_driver_handle_message):
6801 Update for API changes in libdbus.
6803 * dbus/dbus-message.c: (dbus_message_new_reply):
6804 * dbus/dbus-message.h:
6805 Remove the name argument. The spec states that replies shouldn't
6808 2003-02-13 Anders Carlsson <andersca@codefactory.se>
6810 * bus/desktop-file.c: (parse_section_start), (parse_key_value),
6811 (report_error), (bus_desktop_file_load), (lookup_section),
6812 (lookup_line), (bus_desktop_file_get_raw),
6813 (bus_desktop_file_get_string):
6814 * bus/desktop-file.h:
6815 Some fixes, and new functions for getting a key value from a section.
6817 2003-02-13 Havoc Pennington <hp@pobox.com>
6819 * test/data/auth/fail-after-n-attempts.auth-script: new test
6821 * dbus/dbus-auth.c (send_rejected): shutdown_mech() when we
6824 2003-02-13 Havoc Pennington <hp@pobox.com>
6826 * dbus/dbus-auth.c (handle_server_data_external_mech): args to
6827 dbus_credentials_match were backward
6829 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support
6830 NO_CREDENTIALS and ROOT_CREDENTIALS
6832 * dbus/dbus-auth.c (_dbus_auth_do_work): move get_state() routine
6833 into here. Never process more commands after we've reached an
6834 end state; store further data as unused bytes.
6836 * test/data/auth/*: add more auth tests
6838 * dbus/dbus-auth-script.c (_dbus_auth_script_run): support EXPECT
6839 command to match exact string and EXPECT_UNUSED to match unused
6842 * test/Makefile.am (dist-hook): fix to dist all the test stuff
6844 2003-02-12 Havoc Pennington <hp@pobox.com>
6846 * dbus/dbus-string.c (_dbus_string_pop_line): fix to also strip
6847 \r off of popped lines
6849 * dbus/dbus-auth.c (_dbus_auth_test): write code to run auth
6852 * dbus/dbus-auth-script.c (_dbus_auth_script_run): when doing a
6855 2003-02-12 Havoc Pennington <hp@pobox.com>
6857 * dbus/Makefile.am: remove break-loader from the build, since it
6860 * configure.in: add --enable-gcov to turn on coverage profiling
6861 flags and disable optimization
6863 2003-02-10 Havoc Pennington <hp@pobox.com>
6865 * dbus/dbus-auth-script.c, dbus/dbus-auth-script.h: sync
6866 initial cut at test framework for DBusAuth from laptop.
6867 Doesn't quite work yet but it compiles and I need to get
6868 it off the 266mhz laptop. ;-)
6870 * dbus/dbus-server-debug.c (_dbus_server_debug_accept_transport):
6871 fix a memleak in error case
6873 2003-02-12 Anders Carlsson <andersca@codefactory.se>
6876 * bus/desktop-file.c:
6877 * bus/desktop-file.h:
6878 Add a desktop file parser.
6880 2003-02-11 Zack Rusin <zack@kde.org>
6882 * qt/message.[h|cpp]: sample implementation
6883 of the KDE wrapper for DBusMessage
6885 2003-02-09 Zack Rusin <zack@kde.org>
6887 * test/bus-test.c: make_it_compile
6888 * doc/dbus-specification.sgml: minimal semantic fix
6890 2003-02-06 Anders Carlsson <andersca@codefactory.se>
6896 2003-02-06 Anders Carlsson <andersca@codefactory.se>
6899 * dbus/dbus-break-loader.c:
6901 * test/break-loader.c:
6902 Move dbus-break-loader to test/ and rename it to break-loader.
6904 2003-02-02 Havoc Pennington <hp@pobox.com>
6906 * dbus/dbus-keyring.c, dbus/dbus-keyring.h: template files
6907 for code to manage cookies in your home directory
6909 * dbus/dbus-sysdeps.c (_dbus_generate_random_bytes): new function
6911 * dbus/dbus-auth.c (get_state): impose a maximum number of tries
6912 to authenticate, then disconnect the client.
6914 2003-02-03 Alexander Larsson <alexl@redhat.com>
6916 * dbus/dbus-message.c (dbus_message_append_fields):
6919 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6921 * doc/dbus-specification.sgml:
6922 Update address format section.
6924 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6927 * test/bus-test.c: (get_time), (add_timeout), (remove_timeout),
6928 (message_handler), (new_connection_callback), (loop_quit),
6932 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6934 * bus/driver.c: (bus_driver_handle_service_exists):
6935 Simplify the code a bit.
6937 * dbus/dbus-bus.c: (dbus_bus_service_exists):
6940 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6943 Add libdbus-daemon.la and link to it.
6945 2003-02-01 James Willcox <jwillcox@gnome.org>
6947 * bus/driver.c: (bus_driver_handle_own_service):
6948 Actually include the service reply code in the message.
6950 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6952 * bus/driver.c: (bus_driver_handle_service_exists):
6953 Don't unref the incoming message.
6955 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6957 * dbus/dbus.h: Add dbus-address.h and dbus-bus.h
6959 2003-02-02 Anders Carlsson <andersca@codefactory.se>
6961 * dbus/dbus-server.c: (dbus_server_listen):
6962 * dbus/dbus-transport.c: (_dbus_transport_open):
6963 ifdef out the calls to the debug transport and server.
6965 2003-02-02 Alexander Larsson <alexl@redhat.com>
6967 * dbus/dbus-watch.c (dbus_watch_get_flags):
6968 Add note in the docs that ERROR or HANGUP won't be returned
6969 and are assumed always on.
6971 * glib/dbus-gmain.c (add_watch):
6972 Always add IO_ERR | IO_HUP
6974 * dbus/dbus-message.h:
6975 Add semicolon after dbus_message_iter_get_string_array().
6976 Makes qt code build again
6978 2003-02-01 Anders Carlsson <andersca@codefactory.se>
6980 * bus/driver.c: (create_unique_client_name),
6981 (bus_driver_handle_hello):
6982 Don't take a name, just use a numeric id to identify
6986 * dbus/dbus-bus.c: (dbus_bus_register_client),
6987 (dbus_bus_acquire_service), (dbus_bus_service_exists):
6989 Add new convenience functions for communicating with the bus.
6991 * dbus/dbus-message.h:
6993 * dbus/dbus-protocol.h:
6996 2003-02-01 Alexander Larsson <alexl@redhat.com>
6998 * dbus/dbus-message.c (dbus_message_append_fields):
6999 Add some more doc comments.
7001 2003-02-01 Havoc Pennington <hp@pobox.com>
7003 * dbus/dbus-break-loader.c (randomly_modify_length): change
7004 a 4-byte value in the message as if it were a length
7006 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): don't set
7007 execute bit on saved files
7009 2003-02-01 Havoc Pennington <hp@pobox.com>
7011 * dbus/dbus-break-loader.c (main): new program to find messages
7012 that break the loader.
7014 * dbus/dbus-sysdeps.c (_dbus_string_append_uint): new function
7015 * dbus/dbus-sysdeps.c (_dbus_string_save_to_file): new function
7017 * dbus/dbus-string.c (_dbus_string_set_byte): new
7019 2003-01-31 Havoc Pennington <hp@pobox.com>
7021 * dbus/dbus-message.c: refactor the test code to be more general,
7022 in preparation for writing a "randomly permute test cases to
7023 try to break the loader" program.
7025 2003-01-31 Havoc Pennington <hp@pobox.com>
7027 * doc/dbus-specification.sgml: work on the specification
7029 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): check
7030 the protocol version of the message.
7032 * dbus/dbus-protocol.h: drop special _REPLY names, the spec
7033 no longer specifies that.
7034 (DBUS_SERVICE_REPLY_SERVICE_EXISTS): fix flags (1/2/4/8 not
7037 * dbus/dbus-marshal.c (_dbus_marshal_get_arg_end_pos): add missing
7038 "break" for DBUS_TYPE_NIL, remove @todo
7040 2003-01-31 Havoc Pennington <hp@pobox.com>
7042 * dbus/dbus-message.c (dbus_message_set_is_error_reply): rename
7043 just set_is_error/get_is_error as this is a commonly-used
7044 function, and write docs.
7046 2003-01-31 Anders Carlsson <andersca@codefactory.se>
7048 * dbus/dbus-address.c: (dbus_address_entry_free):
7049 Free key and value lists.
7051 * dbus/dbus-internals.c: (_dbus_type_to_string):
7052 Add the types we didn't have.
7054 * dbus/dbus-marshal.c: (_dbus_marshal_get_arg_end_pos),
7055 (_dbus_marshal_validate_arg):
7058 * dbus/dbus-message.c: (dbus_message_set_sender):
7059 Remove todo about being able to set sender to NULL.
7061 (dbus_message_set_is_error_reply),
7062 (dbus_message_get_is_error_reply):
7063 * dbus/dbus-message.h:
7066 * dbus/dbus-protocol.h:
7067 Add error reply flag.
7069 * test/data/valid-messages/opposite-endian.message:
7070 Add NIL type to test.
7072 2003-01-31 Havoc Pennington <hp@pobox.com>
7074 * doc/dbus-specification.sgml: fully specify the header. Add
7075 flags and major protocol version, and change header/body len to
7078 * dbus/dbus-message-builder.c (append_saved_length): append length
7081 * dbus/dbus-message.c (dbus_message_create_header): change header
7082 length and body length to unsigned. Add the new fields from the
7084 (_dbus_message_loader_return_buffer): unsigned header/body len
7086 2003-01-30 Havoc Pennington <hp@pobox.com>
7088 * dbus/dbus-auth.c: rework to use only REJECTED, no
7091 * doc/dbus-sasl-profile.txt: drop MECHANISMS and just
7092 use REJECTED, suggested by Mark McLoughlin
7094 2003-01-30 Havoc Pennington <hp@pobox.com>
7096 * dbus/dbus-server.c (dbus_server_listen): @todo about how we need
7097 a better way to report errors here. e.g. "unix address lacks
7098 path" or something. also "no such file" when the path doesn't
7101 * dbus/dbus-address.c (dbus_address_entries_free): add @todo about
7103 (dbus_parse_address): add @todo about documenting address format,
7104 and allowing , and ; to be escaped
7106 2003-01-30 Anders Carlsson <andersca@codefactory.se>
7109 Add dbus-address.[ch]
7111 * dbus/dbus-address.c: (dbus_address_entry_free),
7112 (dbus_address_entries_free), (create_entry),
7113 (dbus_address_entry_get_method), (dbus_address_entry_get_value),
7114 (dbus_parse_address), (_dbus_address_test):
7115 * dbus/dbus-address.h:
7116 New files for dealing with address parsing.
7118 * dbus/dbus-connection.c:
7119 Document timeout functions.
7121 * dbus/dbus-message.c:
7122 Document dbus_message_new_from_message.
7124 * dbus/dbus-server-debug.c:
7127 * dbus/dbus-server.c: (dbus_server_listen):
7128 Parse address and use correct server implementation.
7130 * dbus/dbus-string.c: (_dbus_string_find_to), (_dbus_string_test):
7131 * dbus/dbus-string.h:
7132 New function with test.
7134 * dbus/dbus-test.c: (dbus_internal_symbol_do_not_use_run_tests):
7138 * dbus/dbus-transport-debug.c:
7141 * dbus/dbus-transport.c: (_dbus_transport_open):
7142 Parse address and use correct transport implementation.
7144 2003-01-30 Havoc Pennington <hp@pobox.com>
7146 * dbus/dbus-message.c: use message->byte_order instead of
7147 DBUS_COMPILER_BYTE_ORDER throughout.
7148 (dbus_message_create_header): pad header to align the
7149 start of the body of the message to 8-byte boundary
7151 * dbus/dbus-marshal.h: make all the demarshalers take const
7152 DBusString arguments.
7154 * dbus/dbus-message.c (_dbus_message_loader_return_buffer):
7155 validate message args here, so we don't have to do slow validation
7156 later, and so we catch bad messages as they are incoming. Also add
7157 better checks on header_len and body_len. Also fill in
7160 * dbus/dbus-string.c (_dbus_string_validate_utf8): new (not
7161 implemented properly)
7162 (_dbus_string_validate_nul): new function to check all-nul
7164 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): rename
7165 get_arg_end_pos and remove all validation
7166 (_dbus_marshal_validate_arg): actually do validation here.
7168 2003-01-29 Havoc Pennington <hp@pobox.com>
7170 * dbus/dbus-message.c (check_message_handling): fix assertion
7171 failure on set_client_serial
7173 2003-01-28 Havoc Pennington <hp@pobox.com>
7175 * dbus/dbus-server-debug.c: Add doc section comments
7177 * dbus/dbus-transport-debug.c: add doc section comments
7179 2003-01-28 Havoc Pennington <hp@redhat.com>
7181 * dbus/dbus-string.c (_dbus_string_base64_decode): append bytes in
7182 the reverse order from how I had it
7183 (_dbus_string_base64_encode): reverse encoding order. I was
7184 basically byteswapping everything during encoding.
7186 2003-01-28 Anders Carlsson <andersca@codefactory.se>
7188 * dbus/dbus-connection-internal.h:
7189 * dbus/dbus-connection.c: (_dbus_connection_add_timeout),
7190 (_dbus_connection_remove_timeout):
7191 Add functions for adding and removing timeouts.
7193 * dbus/dbus-message.c: (dbus_message_new_from_message):
7194 Add new function that takes a message and creates an exact
7195 copy of it, but with the refcount set to 1.
7196 (check_message_handling):
7199 * dbus/dbus-server-protected.h:
7200 * dbus/dbus-server.c: (_dbus_server_init_base),
7201 (_dbus_server_finalize_base), (_dbus_server_add_timeout),
7202 (dbus_server_set_timeout_functions):
7203 (_dbus_server_remove_timeout):
7204 New functions so that a server can add and remove timeouts.
7206 (dbus_server_listen):
7207 Add commented out call to dbus_server_debug_new.
7209 * dbus/dbus-timeout.c: (_dbus_timeout_new):
7210 Actually set the handler, doh.
7212 * dbus/dbus-transport.c: (_dbus_transport_open):
7213 Add commented out call to dbus_transport_debug_client_new.
7216 Add dbus-transport-debug.[ch] and dbus-server-debug.[ch]
7218 2003-01-28 Havoc Pennington <hp@pobox.com>
7220 * dbus/dbus-message.c (check_message_handling): function to check
7221 on the loaded message, iterates over it etc.
7223 2003-01-28 Havoc Pennington <hp@pobox.com>
7225 * test/Makefile.am (dist-hook): fix make distdir
7227 * dbus/Makefile.am (TESTS_ENVIRONMENT): fix make check
7229 2003-01-27 Havoc Pennington <hp@pobox.com>
7231 * dbus/dbus-mempool.c (time_for_size): replace printf with
7234 * dbus/dbus-message-builder.c (_dbus_message_data_load): allow
7235 empty lines; fix the SAVE_LENGTH stuff to be
7236 START_LENGTH/END_LENGTH so it actually works; couple other
7239 * test/Makefile.am (dist-hook): add dist-hook for .message files
7241 * dbus/dbus-string.c (DBUS_STRING_COPY_PREAMBLE): source of a copy
7242 can be constant or locked.
7243 (_dbus_string_free): allow freeing a const string as
7246 * dbus/dbus-sysdeps.c (_dbus_concat_dir_and_file): utility
7248 * dbus/dbus-test-main.c (main): take an argument which is the
7249 directory containing test data
7251 * dbus/dbus-message.c (_dbus_message_test): pass a test_data_dir
7252 argument to this and load all the messages in test/data/
7253 checking that they can be loaded or not loaded as appropriate.
7255 2003-01-27 Anders Carlsson <andersca@codefactory.se>
7257 * bus/dispatch.c: (bus_dispatch_message_handler):
7258 Dispatch messages sent to services.
7260 * bus/driver.c: (bus_driver_send_service_deleted),
7261 (bus_driver_send_service_created), (bus_driver_send_service_lost),
7262 (bus_driver_send_service_acquired):
7263 Add helper functions for sending service related messages.
7265 (bus_driver_send_welcome_message):
7266 Send HELLO_REPLY instead of WELCOME.
7268 (bus_driver_handle_list_services):
7269 Send LIST_SERVICES_REPLY instead of SERVICES.
7271 (bus_driver_handle_own_service),
7272 (bus_driver_handle_service_exists):
7273 New message handlers.
7275 (bus_driver_handle_message):
7276 Invoke new message handlers.
7278 (bus_driver_remove_connection):
7279 Don't remove any services here since that's done automatically
7280 by bus_service_remove_owner now.
7283 New function signatures.
7285 * bus/services.c: (bus_service_add_owner):
7286 Send ServiceAcquired message if we're the only primary owner.
7288 (bus_service_remove_owner):
7289 Send ServiceAcquired/ServiceLost messages.
7291 (bus_service_set_prohibit_replacement),
7292 (bus_service_get_prohibit_replacement):
7293 Functions for setting prohibit replacement.
7295 (bus_service_has_owner):
7296 New function that checks if a connection is in the owner queue of
7300 Add new function signatures.
7302 * dbus/dbus-list.c: (_dbus_list_test):
7303 Add tests for _dbus_list_remove_last and traversing the list backwards.
7306 Fix a typo in _dbus_list_get_prev_link, if we're at the first element we can't
7307 go any further, so return NULL then.
7309 * dbus/dbus-protocol.h:
7310 Add new messages, service flags and service replies.
7312 2003-01-26 Havoc Pennington <hp@pobox.com>
7314 * dbus/dbus-message-builder.c: implement, completely untested.
7316 * test/data/*: add data to be used in testing.
7317 ".message" files are our simple loadable text format.
7318 ".message-raw" will be binary dumps of messages.
7320 * dbus/dbus-string.c (_dbus_string_starts_with_c_str): new
7322 2003-01-26 Havoc Pennington <hp@pobox.com>
7324 * dbus/dbus-sysdeps.c (_dbus_file_get_contents): new function
7326 * dbus/dbus-errors.c (dbus_result_to_string): add
7329 * dbus/dbus-message-builder.c: new file, will contain code to load
7330 up messages from files. Not implemented yet.
7332 2003-01-26 Havoc Pennington <hp@pobox.com>
7334 * dbus/dbus-message.c (dbus_message_set_sender): support deleting
7335 the sender by setting to NULL
7337 2003-01-26 Havoc Pennington <hp@pobox.com>
7339 The unit tests pass, but otherwise untested. If it breaks, the
7340 tests should have been better. ;-)
7342 * bus/driver.c (bus_driver_handle_hello): return if we disconnect
7345 * dbus/dbus-message.c: redo everything so we maintain
7346 message->header as the only copy of the various fields.
7347 This avoids the possibility of out-of-memory in some cases,
7348 for example dbus_message_lock() can't run out of memory anymore,
7349 and avoids extra copying. Figured I may as well go ahead and do
7350 this since it was busted for dbus_message_lock to not return
7351 failure on OOM, and dbus_message_write_header was totally
7352 unchecked for OOM. Also fixed some random other bugs.
7354 * dbus/dbus-marshal.c (_dbus_marshal_get_field_end_pos): verify
7355 that strings are nul-terminated. Also, end_pos can be equal
7356 to string length just not greater than, I think.
7357 (_dbus_marshal_set_int32): new function
7358 (_dbus_marshal_set_uint32): new function
7359 (_dbus_marshal_set_string): new function
7361 * dbus/dbus-connection.c (_dbus_connection_new_for_transport): fix
7362 a warning, init timeout_list to NULL
7363 (dbus_connection_send_message): don't use uninitialized variable
7366 * dbus/dbus-string.c (_dbus_string_replace_len): new function
7368 2003-01-26 Anders Carlsson <andersca@codefactory.se>
7370 * bus/driver.c: (bus_driver_handle_hello),
7371 (bus_driver_send_welcome_message):
7374 2003-01-26 Anders Carlsson <andersca@codefactory.se>
7376 * dbus/dbus-auth.c: (process_auth), (_dbus_auth_unref):
7377 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
7378 (dbus_connection_unref):
7379 * dbus/dbus-marshal.c: (_dbus_marshal_test):
7380 * dbus/dbus-message.c: (dbus_message_unref),
7383 (dbus_message_get_fields):
7384 Remove debugging printout.
7386 (_dbus_message_loader_return_buffer):
7387 Don't store the header string.
7389 (_dbus_message_test):
7392 2003-01-26 Richard Hult <rhult@codefactory.se>
7394 * glib/dbus-gmain.c (dbus_connection_dispatch): Traverse a copy of
7395 the file descriptor list, since it can change under us.
7397 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7399 * glib/dbus-gmain.c: (dbus_connection_prepare),
7400 (dbus_connection_check), (dbus_connection_dispatch), (add_watch),
7401 (remove_watch), (dbus_connection_hookup_with_g_main):
7402 Rewrite the glib handling to use its own GSource instead of a
7403 GIOChannel so we can catch messages put in the queue while waiting
7406 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7409 * bus/connection.c: (connection_disconnect_handler),
7410 (connection_watch_callback), (bus_connection_setup):
7411 * bus/dispatch.c: (send_one_message),
7412 (bus_dispatch_broadcast_message), (bus_dispatch_message_handler),
7413 (bus_dispatch_add_connection), (bus_dispatch_remove_connection):
7415 * bus/driver.c: (bus_driver_send_service_deleted),
7416 (bus_driver_send_service_created), (bus_driver_handle_hello),
7417 (bus_driver_send_welcome_message),
7418 (bus_driver_handle_list_services), (bus_driver_remove_connection),
7419 (bus_driver_handle_message):
7421 Refactor code, put the message dispatching in its own file. Use
7422 _DBUS_HANDLE_OOM. Also send ServiceDeleted messages when a client
7425 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7427 * dbus/dbus-internals.h:
7428 Add _DBUS_HANDLE_OOM macro, it doesn't do anything currently.
7430 * dbus/dbus-message.c: (dbus_message_get_sender):
7431 * dbus/dbus-message.h:
7432 Implement dbus_message_get_sender.
7434 * dbus/dbus-protocol.h:
7435 Add message and service defines.
7437 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7439 * dbus/dbus-connection.c: (dbus_connection_send_message):
7440 * dbus/dbus-message-internal.h:
7441 * dbus/dbus-message.c: (_dbus_message_get_client_serial),
7442 (dbus_message_write_header):
7443 Remove _dbus_messag_unlock and don't set the client serial on a
7444 message if one already exists.
7446 2003-01-24 Havoc Pennington <hp@pobox.com>
7448 * dbus/dbus-list.c (alloc_link): put a thread lock on the global
7451 * bus/driver.c (bus_driver_handle_list_services): fix a leak
7454 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7456 * dbus/dbus-list.c: (alloc_link), (free_link):
7457 Use a memory pool for the links.
7459 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7461 * bus/connection.c: (bus_connection_foreach):
7463 Add new bus_connection_foreach function.
7465 * bus/driver.c: (send_one_message), (bus_driver_broadcast_message):
7466 Add function that broadcasts a message to all clients.
7468 (bus_driver_send_service_created), (bus_driver_handle_hello),
7469 (bus_driver_send_welcome_message),
7470 (bus_driver_handle_list_services), (bus_driver_message_handler):
7471 Implement functions that take care of listing services, and notifying
7472 clients when new services are created.
7474 * bus/services.c: (bus_services_list):
7476 Add new function that returns an array of strings with the currently
7477 registered services.
7480 * glib/dbus-gmain.c:
7481 Update copyright year.
7483 2003-01-25 Anders Carlsson <andersca@codefactory.se>
7485 * dbus/dbus-connection.c: (dbus_connection_send_message):
7486 Unlock the message in case it was sent earlier.
7488 (dbus_connection_send_message_with_reply_and_block):
7489 Remove the reply message from the list.
7491 * dbus/dbus-marshal.c: (_dbus_demarshal_string_array):
7492 Set array_len and new_pos correctly.
7494 (_dbus_marshal_test):
7495 Remove debug output.
7497 * dbus/dbus-message-internal.h:
7498 * dbus/dbus-message.c: (_dbus_message_get_reply_serial):
7499 New function that returns the reply serial.
7501 (_dbus_message_unlock):
7502 New function that unlocks a message and resets its header.
7504 (dbus_message_append_string_array),
7505 (dbus_message_get_fields_valist),
7506 (dbus_message_iter_get_field_type),
7507 (dbus_message_iter_get_string_array),
7508 (dbus_message_get_fields),
7509 (dbus_message_append_fields_valist):
7510 Handle string arrays.
7512 (dbus_message_set_sender):
7513 Make this function public since the bus daemon needs it.
7515 (decode_header_data):
7516 Set the reply serial to -1 initially.
7518 * dbus/dbus-message.h:
7519 Add dbus_message_set_sender.
7521 2003-01-24 Havoc Pennington <hp@pobox.com>
7523 * doc/dbus-specification.sgml: add some stuff
7525 2003-01-22 Havoc Pennington <hp@pobox.com>
7527 * doc/dbus-specification.sgml: Start to document the protocol.
7529 2003-01-22 Havoc Pennington <hp@pobox.com>
7531 * dbus/dbus-connection.c
7532 (dbus_connection_send_message_with_reply_and_block): add some @todo
7534 * bus/driver.c (bus_driver_add_connection): add a FIXME about memleak
7536 2003-01-21 Havoc Pennington <hp@pobox.com>
7538 (patch untested because can't compile)
7540 * bus/driver.c (create_unique_client_name): make this function
7541 never recycle client names. Also, caller should initialize
7544 * dbus/dbus-sysdeps.c (_dbus_get_current_time): new function
7546 2003-01-21 Anders Carlsson <andersca@codefactory.se>
7548 * dbus/dbus-marshal.c: (_dbus_marshal_double),
7549 (_dbus_marshal_int32), (_dbus_marshal_uint32),
7550 (_dbus_marshal_int32_array), (_dbus_marshal_uint32_array),
7551 (_dbus_marshal_double_array), (_dbus_marshal_string_array),
7552 (_dbus_demarshal_int32_array), (_dbus_demarshal_uint32_array),
7553 (_dbus_demarshal_double_array), (_dbus_demarshal_string_array),
7554 (_dbus_marshal_get_field_end_pos), (_dbus_marshal_test):
7555 * dbus/dbus-marshal.h:
7556 * dbus/dbus-protocol.h:
7557 Add support for marshalling and demarshalling integer, double
7560 2003-01-21 Anders Carlsson <andersca@codefactory.se>
7565 * bus/connection.c: (connection_disconnect_handler):
7566 Remove the connection from the bus driver's list.
7568 (connection_watch_callback): Dispatch messages.
7570 (free_connection_data): Free connection name.
7572 (bus_connection_setup): Add connection to the bus driver's list.
7573 (bus_connection_remove_owned_service):
7574 (bus_connection_set_name), (bus_connection_get_name):
7575 Add functions for setting and getting the connection's name.
7578 Add function headers.
7580 * bus/driver.c: (create_unique_client_name),
7581 (bus_driver_handle_hello_message),
7582 (bus_driver_send_welcome_message), (bus_driver_message_handler),
7583 (bus_driver_add_connection), (bus_driver_remove_connection):
7586 * bus/services.c: (bus_service_free):
7588 New file that handles communication and registreation with the bus
7591 2003-01-21 Anders Carlsson <andersca@codefactory.se>
7593 * dbus/dbus-connection.c: (dbus_connection_send_message):
7594 Add a new client_serial parameter.
7596 (dbus_connection_send_message_with_reply):
7597 Remove a @todo since we've implemented the blocking function.
7599 (dbus_connection_send_message_with_reply_and_block):
7600 New function that sends a message and waits for a reply and
7601 then returns the reply.
7603 * dbus/dbus-connection.h:
7606 * dbus/dbus-errors.c: (dbus_result_to_string):
7607 * dbus/dbus-errors.h:
7608 Add new DBUS_RESULT.
7610 * dbus/dbus-message-internal.h:
7611 * dbus/dbus-message.c: (_dbus_message_get_reply_serial),
7612 (_dbus_message_set_sender), (dbus_message_write_header),
7613 (dbus_message_new_reply), (decode_header_data),
7614 (_dbus_message_loader_return_buffer), (_dbus_message_test):
7615 * dbus/dbus-message.h:
7616 Add new functions that set the reply serial and sender.
7617 Also marshal and demarshal them correctly and add test.
7619 * dbus/dbus-protocol.h:
7620 Add new DBUS_MESSAGE_TYPE_SENDER.
7623 * glib/dbus-gmain.c: (watch_callback), (free_callback_data),
7624 (add_watch), (remove_watch), (add_timeout), (remove_timeout),
7625 (dbus_connection_hookup_with_g_main):
7626 * glib/test-dbus-glib.c: (main):
7627 Rewrite to use GIOChannel and remove the GSource crack.
7629 * test/echo-client.c: (main):
7630 * test/watch.c: (check_messages):
7631 Update for changed APIs
7633 2003-01-19 Anders Carlsson <andersca@codefactory.se>
7635 * dbus/Makefile.am: Add dbus-timeout.[cħ]
7637 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport):
7638 Create a DBusTimeoutList.
7639 (dbus_connection_set_timeout_functions): Add new function to
7640 set timeout callbacks
7642 * dbus/dbus-connection.h: Add public DBusTimeout API.
7644 * dbus/dbus-message.c: (dbus_message_get_service):
7645 * dbus/dbus-message.h: New function.
7647 * dbus/dbus-server.c: Fix small doc typo.
7649 * dbus/dbus-timeout.[ch]: New files for mainloop timeouts.
7651 2003-01-19 Anders Carlsson <andersca@codefactory.se>
7653 * dbus/dbus-string.c (_dbus_string_move_len): Don't delete all
7654 of the string, just as long as specified.
7656 2003-01-19 Havoc Pennington <hp@pobox.com>
7658 * dbus/dbus-connection.c (dbus_connection_get_is_authenticated):
7661 * dbus/dbus-server.c (dbus_server_set_max_connections)
7662 (dbus_server_get_max_connections, dbus_server_get_n_connections):
7663 keep track of current number of connections, and add API for
7664 setting a max (but haven't implemented enforcing the max yet)
7666 2003-01-18 Havoc Pennington <hp@pobox.com>
7668 * dbus/dbus-transport-unix.c (unix_do_iteration): only do the
7669 reading/writing if read_watch != NULL or write_watch != NULL.
7671 * dbus/dbus-message.c (_dbus_message_loader_return_buffer): fix
7672 the message loader code to actually load message->header and
7673 message->body into the newly-created message.
7675 * dbus/dbus-transport-unix.c (check_write_watch): fix a mem leak
7678 * dbus/dbus-connection.c (dbus_connection_set_max_message_size)
7679 (dbus_connection_get_max_message_size)
7680 (dbus_connection_set_max_live_messages_size)
7681 (dbus_connection_get_max_live_messages_size): implement some
7682 resource limitation functions
7684 * dbus/dbus-resources.c: new file implementing some of the
7685 resource limits stuff
7687 * dbus/dbus-message.c (dbus_message_iter_get_byte_array): add
7688 missing docs, add @todo to handle OOM etc.
7690 * dbus/dbus-marshal.c (_dbus_demarshal_byte_array): add missing
7693 2003-01-18 Havoc Pennington <hp@pobox.com>
7695 * dbus/dbus-connection.c (dbus_connection_unref): disconnect the
7696 connection if it hasn't been already.
7698 * dbus/dbus-connection.h: kill off the idea of an ErrorFunction,
7699 replace with DisconnectFunction.
7701 2003-01-18 Havoc Pennington <hp@pobox.com>
7703 Building --disable-verbose-mode --disable-asserts --disable-tests
7704 cuts the library from 112K to 45K or so
7706 * configure.in: check for varargs macro support,
7707 add --enable-verbose-mode, --enable-asserts.
7709 * dbus/dbus-internals.h (_dbus_assert): support
7711 (_dbus_verbose): support DBUS_ENABLE_VERBOSE_MODE
7713 2003-01-18 Havoc Pennington <hp@pobox.com>
7715 * dbus/dbus-test.c: include config.h so that tests actually run
7717 * dbus/dbus-string.c: add assertions that stuff is 8-byte aligned,
7718 so the failure mode when that assumption fails will be plenty
7721 2003-01-18 Havoc Pennington <hp@pobox.com>
7723 * configure.in: default --enable-tests to $USE_MAINTAINER_MODE
7725 * dbus/Makefile.am: fix it up so dubs-test-main.c is included in
7728 * test/Makefile.am: don't use special variable "TESTS" for echo-*
7729 since we don't want to use those in make check
7731 2003-01-15 Havoc Pennington <hp@redhat.com>
7737 2003-01-15 Havoc Pennington <hp@redhat.com>
7739 * test/Makefile.am: fix so that test source code ends up in the
7740 distribution on make distcheck
7742 2003-01-15 Havoc Pennington <hp@redhat.com>
7748 2003-01-15 Havoc Pennington <hp@redhat.com>
7750 * dbus/dbus-test.c (dbus_internal_symbol_do_not_use_run_tests):
7751 fix build when --disable-tests
7753 * Makefile.am (EXTRA_DIST): put HACKING in here
7755 * HACKING: document procedure for making a tarball release.
7757 2003-01-14 Anders Carlsson <andersca@codefactory.se>
7759 * bus/connection.c: (connection_error_handler),
7760 (bus_connection_setup):
7761 * bus/main.c: (main):
7762 Make sure that the DBusConnectionData struct is NULLed
7763 out to prevent a segfault.
7765 * dbus/dbus-errors.c: (dbus_result_to_string):
7766 * dbus/dbus-errors.h:
7767 * dbus/dbus-message.c: (dbus_message_get_fields),
7768 (dbus_message_get_fields_valist), (_dbus_message_test):
7769 * dbus/dbus-message.h:
7770 Make dbus_message_get_fields return a result code so we can
7771 track invalid fields as well as oom.
7773 2003-01-11 Havoc Pennington <hp@pobox.com>
7775 * configure.in: change --enable-test/--enable-ansi action-if-given
7776 to enable_foo=$enableval instead of enable_foo=yes
7778 2003-01-08 Havoc Pennington <hp@pobox.com>
7780 * dbus/dbus-string.c (_dbus_string_align_length): new function
7782 * dbus/dbus-test-main.c: move main() for test app here
7784 (dbus_internal_symbol_do_not_use_run_tests): we have to export a
7785 symbol to run tests, because dbus-test isn't in the main
7788 Code review nitpicks.
7790 * dbus/dbus-message.c (dbus_message_write_header): add newlines
7791 for people with narrow emacs ;-). Assert client_serial was filled
7792 in. Assert message->name != NULL.
7793 (dbus_message_append_fields): have "first_field_type" arg separate
7794 from va list, needed for C++ binding that also uses varargs IIRC
7795 and helps with type safety
7796 (dbus_message_new): add @todo about using DBusString to store
7797 service/name internally
7798 (dbus_message_new): don't leak ->service and ->name on OOM later
7800 (dbus_message_unref): free the service name
7801 (dbus_message_get_fields): same change to varargs
7802 i.e. first_field_type
7803 (_dbus_message_loader_return_buffer): assert that the message data
7804 is aligned (if not it's a bug in our code). Put in verbose griping
7805 about why we set corrupted = TRUE.
7806 (decode_header_data): add FIXME that char* is evil. Was going to
7807 add FIXME about evil locale-specific string.h strncmp, but just
7808 switched to wacky string-as-uint32 optimization. Move check for
7809 "no room for field name" above get_const_data_len() to avoid
7810 assertion failure in get_const_data_len if we have trailing 2
7811 bytes or the like. Check for service and name fields being
7812 provided twice. Don't leak service/name on error. Require field
7813 names to be aligned to 4 bytes.
7815 * dbus/dbus-marshal.c: move byte swap stuff to header
7816 (_dbus_pack_int32): uscore-prefix
7817 (_dbus_unpack_int32): uscore-prefix
7818 (_dbus_unpack_uint32): export
7819 (_dbus_demarshal_string): add @todo complaining about use of
7821 (_dbus_marshal_get_field_end_pos): add @todo about bad error
7822 handling allowing corrupt data to go unchecked
7824 2003-01-08 Havoc Pennington <hp@redhat.com>
7826 * dbus/dbus-transport-unix.c (unix_do_iteration): add read/write
7827 to the select() as needed for authentication. (should be using
7828 _dbus_poll() not select, but for another day)
7830 * dbus/dbus.h: include dbus/dbus-protocol.h
7832 2003-01-08 Anders Carlsson <andersca@codefactory.se>
7834 * dbus/Makefile.am (dbusinclude_HEADERS): Install
7837 2003-01-08 Anders Carlsson <andersca@codefactory.se>
7839 * dbus/dbus-internals.c: (_dbus_type_to_string):
7840 New function that returns a string describing a type.
7842 * dbus/dbus-marshal.c: (_dbus_demarshal_byte_array):
7843 * dbus/dbus-marshal.h:
7844 * dbus/dbus-message.c: (dbus_message_get_fields_valist),
7845 (dbus_message_iter_get_field_type), (dbus_message_iter_get_double),
7846 (dbus_message_iter_get_byte_array):
7847 * dbus/dbus-message.h:
7848 Add new convenience functions for appending and getting message fields.
7849 Also add demarshalling routines for byte arrays.
7851 2003-01-07 Anders Carlsson <andersca@codefactory.se>
7853 * dbus/dbus-connection-internal.h:
7854 * dbus/dbus-connection.c: (_dbus_connection_new_for_transport),
7855 (_dbus_connection_get_next_client_serial),
7856 (dbus_connection_send_message):
7857 * dbus/dbus-internals.h:
7858 * dbus/dbus-marshal.c: (unpack_uint32), (dbus_unpack_int32),
7859 (dbus_pack_int32), (_dbus_marshal_double), (_dbus_marshal_int32),
7860 (_dbus_marshal_uint32), (_dbus_demarshal_double),
7861 (_dbus_demarshal_int32), (_dbus_demarshal_uint32),
7862 (_dbus_demarshal_string), (_dbus_marshal_get_field_end_pos),
7863 (_dbus_verbose_bytes), (_dbus_marshal_test):
7864 * dbus/dbus-marshal.h:
7865 * dbus/dbus-message-internal.h:
7866 * dbus/dbus-message.c: (_dbus_message_set_client_serial),
7867 (dbus_message_write_header), (_dbus_message_lock),
7868 (dbus_message_new), (dbus_message_ref), (dbus_message_unref),
7869 (dbus_message_get_name), (dbus_message_append_int32),
7870 (dbus_message_append_uint32), (dbus_message_append_double),
7871 (dbus_message_append_string), (dbus_message_append_byte_array),
7872 (dbus_message_get_fields_iter), (dbus_message_iter_ref),
7873 (dbus_message_iter_unref), (dbus_message_iter_has_next),
7874 (dbus_message_iter_next), (dbus_message_iter_get_field_type),
7875 (dbus_message_iter_get_string), (dbus_message_iter_get_int32),
7876 (dbus_message_iter_get_uint32), (dbus_message_iter_get_double),
7877 (decode_header_data), (_dbus_message_loader_return_buffer),
7878 (message_iter_test), (_dbus_message_test):
7879 * dbus/dbus-message.h:
7880 * dbus/dbus-protocol.h:
7881 * dbus/dbus-test.c: (main):
7883 * glib/test-dbus-glib.c: (message_handler), (main):
7884 * test/echo-client.c: (main):
7885 * test/watch.c: (check_messages):
7886 Make messages sendable and receivable for real.
7888 2003-01-07 Anders Carlsson <andersca@codefactory.se>
7890 * dbus/dbus-marshal.c: (_dbus_marshal_double),
7891 (_dbus_marshal_string), (_dbus_marshal_byte_array):
7892 * dbus/dbus-message.c: (dbus_message_append_int32),
7893 (dbus_message_append_uint32), (dbus_message_append_double),
7894 (dbus_message_append_string), (dbus_message_append_byte_array):
7895 Handle OOM restoration.
7897 2003-01-07 Anders Carlsson <andersca@codefactory.se>
7899 * dbus/dbus-marshal.c: (_dbus_marshal_string),
7900 (_dbus_demarshal_string), (_dbus_marshal_test):
7901 * dbus/dbus-marshal.h:
7902 * dbus/dbus-message.c: (dbus_message_get_name),
7903 Document these functions.
7905 (dbus_message_append_int32), (dbus_message_append_uint32),
7906 (dbus_message_append_double), (dbus_message_append_string),
7907 (dbus_message_append_byte_array):
7908 * dbus/dbus-message.h:
7909 Add functions for adding message fields of different types.
7911 * dbus/dbus-protocol.h:
7912 Add the different types.
7914 2003-01-05 Havoc Pennington <hp@pobox.com>
7916 * bus/connection.c: implement routines for handling connections,
7917 first thing is keeping a list of owned services on each connection
7918 and setting up watches etc.
7920 * bus/services.c: implement a mapping from service names to lists
7923 * dbus/dbus-hash.c: add DBUS_HASH_POINTER
7925 * dbus/dbus-threads.c (dbus_static_mutex_lock): add functions
7926 to use static mutexes for global data
7928 * dbus/dbus-connection.c (dbus_connection_set_data): add new
7929 collection of functions to set/get application-specific data
7930 on the DBusConnection.
7932 2003-01-04 Havoc Pennington <hp@pobox.com>
7934 * dbus/dbus-sysdeps.c (_dbus_sleep_milliseconds): new function
7935 (_dbus_poll): new function
7937 * dbus/dbus-internals.h (_DBUS_STRUCT_OFFSET): new macro
7940 * bus/loop.c: initial code for the daemon main loop
7942 2003-01-04 Havoc Pennington <hp@pobox.com>
7944 * test/watch.c (error_handler): make it safe if the error handler
7945 is called multiple times (if we s/error handler/disconnect
7946 handler/ we should just guarantee it's called only once)
7948 * dbus/dbus-transport.c (_dbus_transport_disconnect): call the
7949 error handler on disconnect (it's quite possible we should
7950 just change the error handler to a "disconnect handler," I'm
7951 not sure we have any other meaningful errors)
7953 * configure.in: check for getpwnam_r
7955 * dbus/dbus-transport.c, dbus/dbus-transport-unix.c,
7956 dbus/dbus-auth.c: add credentials support, add EXTERNAL auth
7957 mechanism as in SASL spec, using socket credentials
7959 * dbus/dbus-sysdeps.c (_dbus_read_credentials_unix_socket): new function
7960 (_dbus_send_credentials_unix_socket): new function
7962 * dbus/dbus-sysdeps.c (_dbus_accept_unix_socket): rename just
7964 (_dbus_write): only check errno if <0 returned
7965 (_dbus_write_two): ditto
7967 2003-01-02 Anders Carlsson <andersca@codefactory.se>
7969 * dbus/dbus-marshal.c: (_dbus_marshal_utf8_string),
7970 (_dbus_marshal_byte_array), (_dbus_demarshal_utf8_string),
7971 (_dbus_marshal_test):
7972 * dbus/dbus-marshal.h:
7973 Add _dbus_marshal_byte_array and rename _dbus_marshal_string
7974 to _dbus_marshal_utf8_string. Also fix some tests.
7976 2002-12-28 Harri Porten <porten@kde.org>
7978 * configure.in: added check for C++ compiler and a very cheesy
7979 check for the Qt integration
7981 * Makefile.am (SUBDIRS): compile qt subdir if support is enabled
7983 * qt/Makefile.am: added
7985 * qt/.cvsignore: added
7987 * qt/dbus-qthread.cc, qt/dbus-qthread.cpp: renamed former to
7988 latter, added #ifdef QT_THREAD_SUPPORT guard.
7990 * dbus/Makefile.am: added missing headers for make dist
7992 2002-12-28 Kristian Rietveld <kris@gtk.org>
7994 * dbus/Makefile.am: fixup export-symbols-regex.
7996 2002-12-27 Anders Carlsson <andersca@codefactory.se>
7998 * acinclude.m4: Add this file and put the
7999 PKG_CHECK_MODULE macro in it.
8001 2002-12-27 Anders Carlsson <andersca@codefactory.se>
8003 * dbus/dbus-marshal.c: (_dbus_marshal_string),
8004 (_dbus_demarshal_double), (_dbus_demarshal_int32),
8005 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
8006 (_dbus_marshal_test):
8007 Make the demarshalling routines align the pos argument.
8008 Add string marshalling tests and fix the obvious bugs
8011 2002-12-26 Havoc Pennington <hp@pobox.com>
8013 * dbus/dbus-auth.c: fixes fixes fixes
8015 * dbus/dbus-transport-unix.c: wire up support for
8016 encoding/decoding data on the wire
8018 * dbus/dbus-auth.c (_dbus_auth_encode_data)
8019 (_dbus_auth_decode_data): append to target string
8020 instead of nuking it.
8022 2002-12-26 Havoc Pennington <hp@pobox.com>
8024 * dbus/dbus-marshal.h (DBUS_COMPILER_BYTE_ORDER): #ifdef
8025 WORDS_BIGENDIAN then compiler byte order is DBUS_BIG_ENDIAN,
8028 * dbus/dbus-marshal.c: Add macros to do int swapping in-place and
8029 avoid swap_bytes() overhead (ignoring possible assembly stuff for
8030 now). Main point is because I wanted unpack_uint32 to implement
8032 (_dbus_verbose_bytes): new function
8034 * dbus/dbus-string.c (_dbus_string_validate_ascii): new function
8036 * dbus/dbus-message.c (_dbus_message_loader_get_is_corrupted): add
8037 mechanism to handle a corrupt message stream
8038 (_dbus_message_loader_new): fix preallocation to only prealloc,
8041 * dbus/dbus-string.c (_dbus_string_skip_blank): fix this function
8042 (_dbus_string_test): enhance tests for copy/move and fix the
8045 * dbus/dbus-transport-unix.c: Hold references in more places to
8046 avoid reentrancy problems
8048 * dbus/dbus-transport.c: ditto
8050 * dbus/dbus-connection.c (dbus_connection_dispatch_message): don't
8051 leak reference count in no-message case
8053 * test/watch.c (do_mainloop): handle adding/removing watches
8054 during iteration over the watches. Also, ref the connection/server
8055 stored on a watch, so we don't try to mangle a destroyed one.
8057 * dbus/dbus-transport-unix.c (do_authentication): perform
8060 * dbus/dbus-auth.c (get_state): add a state
8061 AUTHENTICATED_WITH_UNUSED_BYTES and return it if required
8062 (_dbus_auth_get_unused_bytes): append the unused bytes
8063 to the passed in string, rather than prepend
8065 * dbus/dbus-transport.c (_dbus_transport_init_base): create
8066 the auth conversation DBusAuth
8068 * dbus/dbus-transport-unix.c (_dbus_transport_new_for_fd)
8069 (_dbus_transport_new_for_domain_socket): when creating a
8070 transport, pass in whether it's a client-side or server-side
8071 transport so we know which DBusAuth to create
8073 2002-12-03 Havoc Pennington <hp@pobox.com>
8075 * dbus/dbus-transport-unix.c (unix_finalize): finalize base
8076 _after_ finalizing the derived members
8077 (unix_connection_set): unref watch if we fail to add it
8079 * dbus/dbus-connection.c (dbus_connection_unref): delete the
8080 transport first, so that the connection owned by the
8081 transport will be valid as the transport finalizes.
8083 * dbus/dbus-transport-unix.c (unix_finalize): free the write_watch
8084 if necessary, and remove watches from the connection.
8086 * dbus/dbus-watch.c (_dbus_watch_list_free): improve a comment
8088 2002-12-26 Anders Carlsson <andersca@codefactory.se>
8090 * dbus/dbus-marshal.c: (_dbus_marshal_string),
8091 (_dbus_demarshal_double), (_dbus_demarshal_int32),
8092 (_dbus_demarshal_uint32), (_dbus_demarshal_string),
8093 (_dbus_marshal_test):
8094 * dbus/dbus-marshal.h:
8095 Add string marshal functions and have the demarshal functions
8096 return the new position.
8098 2002-12-25 Havoc Pennington <hp@pobox.com>
8100 * doc/dbus-sasl-profile.txt: docs on the authentication protocol,
8101 it is a simple protocol that just maps directly to SASL.
8103 * dbus/dbus-auth.h, dbus/dbus-auth.c: authentication protocol
8104 initial implementation, not actually used yet.
8106 * dbus/dbus-string.c (_dbus_string_find): new function
8107 (_dbus_string_equal): new function
8108 (_dbus_string_base64_encode): new function
8109 (_dbus_string_base64_decode): new function
8111 2002-12-25 Anders Carlsson <andersca@codefactory.se>
8114 * dbus/dbus-marshal.c: (swap_bytes), (_dbus_marshal_double),
8115 (_dbus_marshal_int32), (_dbus_marshal_uint32),
8116 (_dbus_demarshal_double), (_dbus_demarshal_int32),
8117 (_dbus_demarshal_uint32), (_dbus_marshal_test):
8118 * dbus/dbus-marshal.h:
8119 * dbus/dbus-protocol.h:
8120 * dbus/dbus-test.c: (main):
8122 Add un-optimized marshalling/demarshalling routines.
8124 2002-12-25 Harri Porten <porten@kde.org>
8126 * qt/dbus-qt.h: adjusted ctor and getter to KDE/Qt conventions
8128 2002-12-24 Zack Rusin <zack@kde.org>
8130 * qt/dbus-qthread.cc: adding - integrates QMutex into Dbus
8131 * qt/dbus-qt.h: skeleton with two sample implemenatation of the
8134 2002-12-24 Havoc Pennington <hp@pobox.com>
8136 * glib/dbus-gthread.c: fix include
8138 * glib/dbus-glib.h: rename DBusMessageHandler for now.
8139 I think glib API needs to change, though, as you don't
8140 want to use DBusMessageFunction, you want to use the
8141 DBusMessageHandler object. Probably
8142 dbus_connection_open_with_g_main_loop()
8143 and dbus_connection_setup_g_main_loop() or something like that
8144 (but think of better names...) that just create a connection
8145 that has watch/timeout functions etc. already set up.
8147 * dbus/dbus-connection.c
8148 (dbus_connection_send_message_with_reply): new function just to
8149 show how the message handler helps us deal with replies.
8151 * dbus/dbus-list.c (_dbus_list_remove_last): new function
8153 * dbus/dbus-string.c (_dbus_string_test): free a string that
8156 * dbus/dbus-hash.c: use memory pools for the hash entries
8157 (rebuild_table): be more paranoid about overflow, and
8158 shrink table when we can
8159 (_dbus_hash_test): reduce number of sprintfs and write
8160 valid C89. Add tests for case where we grow and then
8161 shrink the hash table.
8163 * dbus/dbus-mempool.h, dbus/dbus-mempool.c: memory pools
8165 * dbus/dbus-connection.c (dbus_connection_register_handler)
8166 (dbus_connection_unregister_handler): new functions
8168 * dbus/dbus-message.c (dbus_message_get_name): new
8170 * dbus/dbus-list.c: fix docs typo
8172 * dbus/dbus-message-handler.h, dbus/dbus-message-handler.c:
8173 an object representing a handler for messages.
8175 2002-12-16 Anders Carlsson <andersca@codefactory.se>
8178 * glib/dbus-gthread.c: (dbus_gthread_init):
8179 Don't use the gdbus prefix for public functions.
8181 2002-12-16 Anders Carlsson <andersca@codefactory.se>
8185 Add GLib checks and fixup .pc files
8189 * glib/dbus-gmain.c: (gdbus_connection_prepare),
8190 (gdbus_connection_check), (gdbus_connection_dispatch),
8191 (gdbus_add_connection_watch), (gdbus_remove_connection_watch),
8192 (dbus_connection_gsource_new):
8193 * glib/dbus-gthread.c: (dbus_gmutex_new), (dbus_gmutex_free),
8194 (dbus_gmutex_lock), (dbus_gmutex_unlock), (dbus_gthread_init):
8195 * glib/test-dbus-glib.c: (message_handler), (main):
8198 2002-12-15 Harri Porten <porten@kde.org>
8200 * autogen.sh: check for libtoolize before attempting to use it
8202 * dbus/dbus-transport-unix.c: include <sys/time.h> for timeval
8205 * .cvsignore: ignore more stamp files
8207 * dbus/dbus-watch.c (_dbus_watch_list_new): fixed doc error
8209 * test/Makefile.am: added -I$(top_srcdir) to be able to compile
8210 without make install.
8212 2002-12-15 Havoc Pennington <hp@pobox.com>
8214 * dbus/dbus-threads.c: add thread stubs that a higher library
8215 layer can fill in. e.g. the GLib wrapper might fill them in with
8216 GThread stuff. We still need to use this thread API to
8217 thread-safe-ize the library.
8219 2002-12-12 Havoc Pennington <hp@pobox.com>
8221 * dbus/dbus-transport-unix.c, dbus/dbus-server-unix.c: use the
8222 below new interfaces and include fewer system headers.
8224 * dbus/dbus-sysdeps.c (_dbus_read): new function
8225 (_dbus_write): new function
8226 (_dbus_write_two): new function
8227 (_dbus_connect_unix_socket): new function
8228 (_dbus_listen_unix_socket): new function
8230 * dbus/dbus-message-internal.h: change interfaces to use
8233 2002-12-11 Havoc Pennington <hp@pobox.com>
8235 * dbus/dbus-types.h: add dbus_unichar
8237 * dbus/dbus-internals.c (_dbus_verbose): use _dbus_getenv
8239 * dbus/dbus-connection.c (dbus_connection_send_message): return
8242 * dbus/dbus-transport.c: include dbus-watch.h
8244 * dbus/dbus-connection.c: include dbus-message-internal.h
8246 * HACKING: add file with coding guidelines stuff.
8248 * dbus/dbus-string.h, dbus/dbus-string.c: Encapsulate all string
8249 handling here, for security purposes (as in vsftpd). Not actually
8250 using this class yet.
8252 * dbus/dbus-sysdeps.h, dbus/dbus-sysdeps.c: Encapsulate all
8253 system/libc usage here, as in vsftpd, for ease of auditing (and
8254 should also simplify portability). Haven't actually moved all the
8255 system/libc usage into here yet.
8257 2002-11-25 Havoc Pennington <hp@pobox.com>
8259 * dbus/dbus-internals.c (_dbus_verbose): fix to not
8260 always print the first verbose message.
8262 2002-11-24 Havoc Pennington <hp@pobox.com>
8264 * test/echo-client.c, test/echo-server.c: cheesy test
8267 * configure.in (AC_CHECK_FUNCS): check for writev
8269 * dbus/dbus-message.c (_dbus_message_get_network_data): new
8272 * dbus/dbus-list.c (_dbus_list_foreach): new function
8274 * dbus/dbus-internals.c (_dbus_verbose): new function
8276 * dbus/dbus-server.c, dbus/dbus-server.h: public object
8277 representing a server that listens for connections.
8279 * dbus/.cvsignore: create
8281 * dbus/dbus-errors.h, dbus/dbus-errors.c:
8282 public API for reporting errors
8284 * dbus/dbus-connection.h, dbus/dbus-connection.c:
8285 public object representing a connection that
8286 sends/receives messages. (Same object used for
8287 both client and server.)
8289 * dbus/dbus-transport.h, dbus/dbus-transport.c:
8290 Basic abstraction for different kinds of stream
8291 that we might read/write messages from.
8293 2002-11-23 Havoc Pennington <hp@pobox.com>
8295 * dbus/dbus-internals.h (_DBUS_INT_MAX): add _DBUS_INT_MIN
8298 * dbus/dbus-test.c (main): add list test, and include
8299 dbus-test.h as intended
8301 * dbus/dbus-hash.c (_dbus_hash_table_remove_string)
8302 (_dbus_hash_table_remove_int): return value indicates
8303 whether the entry existed to remove
8305 * dbus/dbus-list.c: add linked list utility class,
8308 * dbus/dbus-hash.c: add TODO item about shrinking the hash bucket
8311 2002-11-23 Havoc Pennington <hp@pobox.com>
8313 * Doxyfile.in (INCLUDE_FILE_PATTERNS): expand DBUS_BEGIN_DECLS/
8314 DBUS_END_DECLS to nothing, that should fix this once and for all
8316 * Doxyfile.in (JAVADOC_AUTOBRIEF): set to YES
8318 * dbus/dbus-message.c, dbus/dbus-hash.c:
8319 add some missing @brief
8321 2002-11-23 Havoc Pennington <hp@pobox.com>
8323 * dbus/dbus-message.h: put semicolons after DEBUG_BEGIN_DECLS
8324 to avoid confusing Doxygen
8326 * dbus/dbus-hash.c: @} not }@
8328 * dbus/dbus-message.c (struct DBusMessage): split out
8331 2002-11-23 Havoc Pennington <hp@pobox.com>
8333 * configure.in: pile on more warning flags if using gcc
8335 * Doxyfile.in (EXTRACT_STATIC): set to NO, so we don't have
8336 to document static functions
8338 * configure.in: add summary to end of configure so it
8339 looks nice and attractive
8341 * dbus/dbus-hash.c: finish implementation and write unit
8344 * configure.in: add --enable-tests to enable unit tests
8346 * dbus/dbus-test.c: test program to run unit tests
8347 for all files in dbus/*, initially runs a test for
8350 * dbus/dbus-internals.h: file to hold some internal utility stuff
8352 2002-11-22 Havoc Pennington <hp@redhat.com>
8354 * dbus/dbus-hash.c: copy in Tcl hash table, not yet
8355 "ported" away from Tcl
8357 * dbus/dbus-types.h: header for types such as dbus_bool_t
8359 2002-11-22 Havoc Pennington <hp@redhat.com>
8361 * dbus/dbus.h: fixups for doc warnings
8363 * Doxyfile.in (FILE_PATTERNS): we need to scan .h to pick up
8365 (QUIET): make it quiet so we can see warnings
8367 * dbus/dbus-memory.c: teach D-BUS to allocate and free memory
8369 2002-11-22 Havoc Pennington <hp@redhat.com>
8371 * Makefile.am: include "Doxyfile" target in all-local
8373 * configure.in: generate the Doxyfile
8375 * Doxyfile.in: move Doxyfile here, so we can use
8376 configure to generate a Doxyfile with the right
8379 2002-11-22 Havoc Pennington <hp@redhat.com>
8381 * dbus/dbus-message.c: move inline docs into .c file
8383 * Doxyfile (OUTPUT_DIRECTORY): move output to doc/api
8384 so all docs are under doc/
8385 (MAN_EXTENSION): generate man pages. Use extension
8386 ".3dbus" which matches ".3qt" on my system,
8387 I guess this is OK, I don't know really.
8388 (FILE_PATTERNS): look for .c files not .h, makes sense
8391 2002-11-22 Havoc Pennington <hp@pobox.com>
8393 * Makefile.am (SUBDIRS): rename subdir "server" to "bus"
8394 because any app can be a server, and any app can be a client,
8395 the bus is a special kind of server.
8397 Thu Nov 21 23:35:31 2002 Zack Rusin <zack@kde.org>
8399 * Doxyfile : adding. Still needs Makefile rules to be generated
8400 automatically (just run "doxygen" in the toplevel dir for now to
8403 * dbus/dbus-message.h : Adding sample docs (javadoc since
8404 resembles gtk-doc a little more)
8406 * dbus/dbus.h : Adding sample docs
8408 2002-11-21 Havoc Pennington <hp@redhat.com>
8410 * dbus/Makefile.am (INCLUDES): define DBUS_COMPILATION
8411 so we can allow ourselves to include files directly,
8412 instead of having to use dbus.h
8414 * dbus/dbus.h: fill in
8416 * dbus/dbus-message.h: sketch out a sample header file.
8417 Include griping if you include it directly instead of
8420 * dbus/dbus-macros.h: new file with macros for extern "C",
8421 TRUE/FALSE, NULL, etc.
8423 * doc/file-boilerplate.c: put include guards in here
8425 2002-11-21 Havoc Pennington <hp@redhat.com>
8427 * doc/file-boilerplate.c: include both AFL and GPL boilerplate.
8429 * COPYING: include the GPL as well, and license code
8430 under both AFL and GPL.
8432 2002-11-21 Havoc Pennington <hp@redhat.com>
8434 * acconfig.h: get rid of this
8436 * autogen.sh (run_configure): add --no-configure option
8438 * configure.in: remove AC_ARG_PROGRAM to make
8439 autoconf complain less. add AC_PREREQ.
8440 add AC_DEFINE third arg.
8442 2002-11-21 Anders Carlsson <andersca@codefactory.se>
8445 Fix references so we can distcheck.
8447 2002-11-21 Havoc Pennington <hp@redhat.com>
8449 * Initial module creation