1: 2006-12-12 John (J5) Palmieri <johnp@redhat.com> 2: 3: * Released 1.0.2 4: 5: 2006-12-12 John (J5) Palmieri <johnp@redhat.com> 6: 7: * bus/signal.c: Fix match_rule_equal errata 8: (CVE-2006-6107 - Patch from Kimmo Hämäläinen 9: <kimmo.hamalainen@nokia.com>) 10: 11: 2006-11-19 Thiago Macieira <thiago@kde.org> 12: 13: * dbus/dbus-sysdeps-pthread.c (_dbus_pthread_mutex_lock, 14: _dbus_pthread_condvar_wait, 15: _dbus_pthread_condvar_wait_timeout): set pmutex->holder to 16: pthread_self() after coming back from a conditional variable 17: wait as well as in one codepath where it was forgotten. 18: Approved by: Havoc Pennington. 19: 20: 2006-11-17 Havoc Pennington <hp@redhat.com> 21: 22: * update-dbus-docs.sh: allow setting fd.org username via env 23: variable. Make it run autogen with --enable-xml-docs=yes 24: --enable-doxygen-docs=yes so configure will fail if the required 25: tools are missing. 26: 27: 2006-11-17 Havoc Pennington <hp@redhat.com> 28: 29: * doc/dbus-faq.xml: minor FAQ tweaks 30: 31: 2006-11-17 John (J5) Palmieri <johnp@redhat.com> 32: 33: * Released 1.0.1 34: 35: 2006-11-14 Havoc Pennington <hp@redhat.com> 36: 37: * dbus/dbus-misc.c, dbus/dbus-misc.h: Move 38: dbus_get_local_machine_id() to its own file, no substantive 39: changes. There are a couple other things we might want to add that 40: are "misc" so moving out of dbus-connection.[hc] which is big 41: enough already. 42: 43: 2006-11-14 Havoc Pennington <hp@redhat.com> 44: 45: * dbus/dbus-internals.c (_dbus_generate_uuid): The spec said the 46: UUID had the timestamp last, but the implementation had it first; 47: move it to last since I think it's a tiny bit nicer (easier to 48: compare at a glance, faster to sort, less code), and will not 49: cause any practical compatibility problems. Also, always convert 50: the timestamp to big endian. 51: 52: * doc/dbus-specification.xml: Clean up the docs on the UUID. 53: 54: * tools/dbus-uuidgen.1: more prominently say it is not suitable 55: as a replacement for regular uuidgen/RFC4122. 56: 57: 2006-11-14 John (J5) Palmieri <johnp@redhat.com> 58: 59: * configure.in: Bump version to 1.0.1 for next stable release 60: 61: * dbus/dbus-threads.h: fix DBUS_THREAD_FUNCTIONS_ALL_MASK to have 62: the correct value so we don't assert when initalizing recursive threads 63: 64: * test/name-test/test-thread-init.c: call dbus_threads_init_default 65: instead of _dbus_threads_init_debug since it is more of a real world 66: test 67: 68: 2006-11-09 John (J5) Palmieri <johnp@redhat.com> 69: 70: * Released 1.0.0 71: 72: 2006-11-09 John (J5) Palmieri <johnp@redhat.com> 73: 74: * AUTHORS: added Peter and Tor of windows porting fame 75: 76: 2006-11-08 Havoc Pennington <hp@redhat.com> 77: 78: * doc/dbus-specification.xml: add a note about protocol stability 79: 80: * doc/TODO: Remove "important for 1.0" section, replace with 81: "important for 1.2" section 82: 83: 2006-11-08 John (J5) Palmieri <johnp@redhat.com> 84: 85: * README: remove some of the out of date flags 86: 87: * doc/TODO: Remove todo items which are done 88: 89: 2006-11-08 John (J5) Palmieri <johnp@redhat.com> 90: 91: * tools/dbus-monitor.c: Handle sigint so we flush stdout 92: (Patch from Olli Salli <olli dot salli at collabora dot co dot uk>) 93: 94: 2006-11-08 John (J5) Palmieri <johnp@redhat.com> 95: 96: * tools/dbus-monitor.1: update the manpage for dbus-monitor 97: (Patch from Olli Salli <olli dot salli at collabora dot co dot uk>) 98: 99: 2006-11-08 John (J5) Palmieri <johnp@redhat.com> 100: 101: * AUTHORS: Update with all people listed in the ChangeLog 102: 103: 2006-11-08 John (J5) Palmieri <johnp@redhat.com> 104: 105: * tools/dbus-monitor.c: add profiling tools to dbus-monitor 106: (Patch from Olli Salli <olli dot salli at collabora dot co dot uk>) 107: 108: 2006-11-07 Havoc Pennington <hp@redhat.com> 109: 110: * HACKING: Update release instructions to include stuff about 111: stable releases, branching, etc. May not be totally correct, 112: please fix if needed, but keep instructions up-to-date so we do 113: each stable release consistently in the future. 114: 115: 2006-11-07 Havoc Pennington <hp@redhat.com> 116: 117: * doc/dbus-specification.xml, doc/dbus-faq.xml, README: various 118: documentation updates. Bump faq/spec versions (not to 1.0; I don't 119: think the spec will be "finished"/1.0 when we ship the 1.0 library). 120: 121: 2006-11-06 John (J5) Palmieri <johnp@redhat.com> 122: 123: * bus/bus.c: Missed patch - pass in the context to the directory watch 124: 125: 2006-11-02 John (J5) Palmieri <johnp@redhat.com> 126: 127: * Released D-Bus 1.0 RC 3(0.95) 128: 129: 2006-11-02 John (J5) Palmieri <johnp@redhat.com> 130: 131: * bus/activation.c (bus_activation_new): don't fail if we can not read 132: the directory as it may not exist 133: 134: 2006-11-02 John (J5) Palmieri <johnp@redhat.com> 135: 136: * bus/config-parser.c (service_dirs_find_dir): use 137: _dbus_list_get_next_link so we don't get stuck in an infinite loop 138: (start_busconfig_child): move processing of standard_session_servicedirs 139: tags here because they have no content 140: (bus_config_parser_content): check we don't have content in 141: standard_session_servicedirs tag 142: 143: 2006-11-02 John (J5) Palmieri <johnp@redhat.com> 144: 145: * tools/Makefile.am: Make sure the /var/lib/dbus directory is created 146: Packagers need to own this directory 147: 148: 2006-11-02 John (J5) Palmieri <johnp@redhat.com> 149: 150: * tools/run-with-tmp-session-bus.sh: fixed script to replace the 151: <standard_session_servicedirs> tag with a <sevicedir> tag that 152: points to the test service directory 153: 154: 2006-11-02 John (J5) Palmieri <johnp@redhat.com> 155: 156: * configure.in: Add a check for DIR *dirp->dd_fd 157: 158: * dbus/dbus-sysdeps-util-unix: if HAVE_DDFD is defined use 159: DIR *dirp->dd_fd else use DIR *dirp->__dd_fd. This takes care 160: of both Solaris and Irix 161: 162: 2006-11-01 Havoc Pennington <hp@redhat.com> 163: 164: * bus/dbus-daemon.1.in: document standard_session_servicedirs 165: 166: 2006-11-01 John (J5) Palmieri <johnp@redhat.com> 167: 168: * configure.in: expose DBUS_DATADIR 169: 170: * bus/config-parser.c: add the standard_session_servicedirs element 171: to the parser 172: (bus_config_parser_content): process the standard_session_servicedirs 173: element by getting the standard directories from sysdeps and merging 174: them into the service directory list 175: (test_default_session_servicedirs): make sure we get what we expect 176: 177: * bus/session.conf.in: replace the servicedir tag with the 178: standard_session_servicedirs tag 179: 180: * dbus/dbus-list.h: remove the typedef of DBusList and place it in 181: dbus-sysdeps.h to avoid circular header dependencies 182: 183: * dbus/dbus-sysdeps.h: add the typedef of DBusList 184: 185: * dbus/dbus-sysdeps-unix.c (split_paths_and_append): utility function 186: which takes a string of directories delimited by colons, parses them 187: out, appends a suffix and puts them in a list ignoring empty elements 188: (_dbus_get_standard_session_servicedirs): returns the standard 189: directories for a session bus to look for service activation files 190: on Unix which includes the XDG_DATA_HOME, XDG_DATA_DIRS and 191: DBUS_DATADIR directories 192: 193: * test/data/valid-config-files/many-rules.conf: add the 194: standard_session_servicedirs tag to the valid config file tests 195: 196: 2006-10-30 Havoc Pennington <hp@redhat.com> 197: 198: * tools/dbus-launch.1, doc/TODO: capture intent to change the 199: command line options of dbus-launch a bit post 1.0 200: 201: 2006-10-27 Havoc Pennington <hp@redhat.com> 202: 203: * dbus/dbus-connection.c (dbus_connection_open_private): document 204: how to choose shared vs. private connections 205: 206: 2006-10-27 Havoc Pennington <hp@redhat.com> 207: 208: * dbus/dbus-test.c: enclose more of the file in the 209: DBUS_BUILD_TESTS check. 210: 211: * dbus/dbus-sysdeps-pthread.c (PTHREAD_CHECK): fix for 212: DBUS_DISABLE_ASSERT case. 213: 214: * dbus/dbus-connection.c (dbus_connection_get_unix_user): document 215: that it only works on the server side 216: 217: * dbus/dbus-bus.c: add a global lock covering the BusData we 218: attach to each connection 219: (internal_bus_get): lock our access to the BusData 220: (dbus_bus_register): lock the entire registration process 221: with _DBUS_LOCK(bus_datas). If we get the lock and 222: registration is already complete, silently return (vs. previous 223: behavior of aborting). 224: (dbus_bus_set_unique_name): lock the BusData 225: (dbus_bus_get_unique_name): lock the BusData 226: 227: 2006-10-27 John (J5) Palmieri <johnp@redhat.com> 228: 229: * bus/config-parser.c (service_dirs_find_dir, 230: service_dirs_append_unique_or_free, 231: service_dirs_append_link_unique_or_free): New static methods 232: for only appending unique service directory names into 233: the service directory list 234: (merge_included, bus_config_parser_content): Only add unique 235: service directory names into the list 236: 237: 2006-10-27 Havoc Pennington <hp@redhat.com> 238: 239: * dbus/dbus-sysdeps-pthread.c: make the "count" and "holder" 240: variables volatile, suggested by Thiago. Document struct fields. 241: (PTHREAD_CHECK): remove pthread error checking if assertions are 242: disabled, should reduce the no-assertions case to the bare 243: minimum code. 244: 245: 2006-10-26 Havoc Pennington <hp@redhat.com> 246: 247: * dbus/dbus-sysdeps-pthread.c (_dbus_pthread_mutex_lock): change 248: to be recursive 249: (_dbus_pthread_mutex_unlock): make it recursive 250: (_dbus_pthread_condvar_wait): save/restore the recursion count 251: (_dbus_pthread_condvar_wait_timeout): save/restore the recursion count 252: 253: 2006-10-26 Havoc Pennington <hp@redhat.com> 254: 255: * doc/dbus-specification.xml: clarify the UUID text slightly 256: 257: * dbus/dbus-sysdeps-pthread.c: check for and mostly abort on 258: pthread errors. Add DBusMutexPThread and DBusCondVarPThread 259: in preparation for being able to extend them for e.g. recursive 260: mutexes. 261: 262: 2006-10-26 Havoc Pennington <hp@redhat.com> 263: 264: * dbus/dbus-threads.[hc]: Documentation improvements. Clarify how 265: condition variables relate to recursive mutexes. 266: 267: * dbus/dbus-sysdeps-pthread.c, dbus/dbus-sysdeps-win-thread.c, 268: dbus/dbus-threads.c: Split the platforms-specific thread 269: implementations into their own files. 270: 271: * dbus/dbus-sysdeps-pthread.c 272: (_dbus_pthread_condvar_wait_timeout): invert the return value, it 273: was backward. Not that anything uses it. 274: 275: 2006-10-26 John (J5) Palmieri <johnp@redhat.com> 276: 277: * dbus-sysdeps-unix.c (_dbus_set_local_creds): Clean up the 278: LOCAL_CREDS vs CMSGCRED stuff a bit. Prefer CMSGCRED. This 279: needs to be cleaned up more. 280: 281: * doc/TODO: Add a todo that *BSD hackers should look at cleaning 282: up the CREDS issue. 283: 284: 2006-10-26 John (J5) Palmieri <johnp@redhat.com> 285: 286: * configure.in, dbus-1.pc.in: Check to see if thread methods 287: are in glibc or libpthread and add -lpthread to the link stage 288: if it is the latter 289: 290: 2006-10-26 Thiago Macieira <thiago@kde.org> 291: 292: * dbus/dbus-connection.c (_dbus_connection_open_internal): Fix 293: bug 8780: the connection lock is only required while recording 294: the shared connection, so protect only that code 295: section. Don't require connection_lookup_shared to return a 296: locked connection. 297: 298: 2006-10-26 Thiago Macieira <thiago@kde.org> 299: 300: * tools/dbus-launch-x11.c (get_session_file, init_x_atoms): 301: check if get_machine_uuid() returns NULL before proceeding any 302: further: we can't init the X atoms or create a session file 303: name if there is no machine ID. 304: This solves a crash reported by some users if 305: --exit-with-session was used without --autolaunch=<machine-id> 306: 307: 2006-10-26 John (J5) Palmieri <johnp@redhat.com> 308: 309: * tools/dbus-launch.c (main): run the dbus-daemon in the builddir 310: if tests are enabled and the DBUS_USE_TEST_BINARY env variable is set 311: 312: * tools/run-with-tmp-session-bus.sh: set DBUS_USE_TEST_BINARY 313: before we run dbus-launch 314: 315: * configure.in: define TEST_BUS_BINARY to be the full path to 316: dbus-daemon in the build root 317: 318: 2006-10-25 Sjoerd Simons <sjoerd@luon.net> 319: 320: * dbus/Makefile.am: 321: * tools/Makefile.am: 322: * toos/dbus-uuidgen.1: 323: Move the machine-id file to /var/lib/dbus 324: 325: 2006-10-24 David Zeuthen <davidz@redhat.com> 326: 327: * dbus/dbus-threads.c (init_uninitialized_locks): Fix typo so it's 328: _dbus_assert (thread_init_generation != _dbus_current_generation) 329: not _dbus_assert (thread_init_generation == _dbus_current_generation) 330: 331: 2006-10-24 Thiago Macieira <thiago@kde.org> 332: 333: * dbus/dbus-sysdeps.h: 334: * dbus/dbus-sysdeps-unix.c: Add function 335: _dbus_make_file_world_readable that chmods a file to 0644. 336: 337: * dbus/dbus-sysdeps-unix.c (_dbus_get_autolaunch_address): 338: Avoid writing to the static "argv" array, thereby avoiding a 339: COW on the child process. 340: 341: * dbus/dbus-internals.c (_dbus_create_uuid_file_exclusively): 342: call _dbus_make_file_world_readable on the created file. 343: 344: 2006-10-23 David Zeuthen <davidz@redhat.com> 345: 346: * dbus/dbus-memory.c: Use atomic variable to protect 347: n_blocks_outstanding otherwise OOM will be reported using SMP on 348: some arches 349: 350: * bus/dispatch.c: Add missing end of line characters 351: 352: * bus/desktop-file.c (parse_section_start, parse_key_value) 353: (bus_desktop_file_load): Propertly handle OOM 354: 355: * dbus/dbus-threads.c (init_uninitialized_locks): Check that 356: thread_init_generation equals _dbus_current_generation, not 0 357: 358: 2006-10-22 Havoc Pennington <hp@redhat.com> 359: 360: * dbus/dbus-connection-internal.h: move prototype of 361: _dbus_bus_notify_shared_connection_disconnected_unlocked() here so 362: it isn't in a public header and doesn't end up in the DBusBus 363: group in Doxygen 364: 365: 2006-10-21 Havoc Pennington <hp@redhat.com> 366: 367: * Makefile.am (EXTRA_DIST): dist cleanup-man-pages.sh so it's in 368: the tarball if packagers want to run it 369: 370: * cleanup-man-pages.sh: Add a script which munges all the internal 371: API man pages out of the Doxygen output. This reduces the size of 372: the installed man pages from 7 to 2 megs, and avoids 373: namespace-polluting pages. Right now (like Doxygen) this script 374: isn't in the build, it's something packagers can do manually. 375: 376: 2006-10-21 Havoc Pennington <hp@redhat.com> 377: 378: * dbus/dbus-connection.h: don't include the watch/timeout 379: functions in connection section 380: 381: 2006-10-21 Havoc Pennington <hp@redhat.com> 382: 383: * Documentation! Whee! Doxygen now 100% silent. If you make it 384: angry again, you will be punished. 385: 386: 2006-10-21 Havoc Pennington <hp@redhat.com> 387: 388: * More documentation - all public API now documented according to 389: Doxygen 390: 391: 2006-10-21 Havoc Pennington <hp@redhat.com> 392: 393: * Document a bunch of undocumented stuff 394: 395: 2006-10-21 Havoc Pennington <hp@redhat.com> 396: 397: * Clean up Doxygen group markers for public API so Doxygen finds 398: everything (not comprehensively fixed for private API). 399: Means all remaining Doxygen warnings are just about missing docs 400: and thus pretty simple to resolve. 401: 402: 2006-10-21 Havoc Pennington <hp@redhat.com> 403: 404: * dbus/dbus-macros.h: add DBUS_GNUC_DEPRECATED macro 405: 406: * dbus/dbus-message.h: mark dbus_message_iter_get_array_len() 407: as DBUS_GNUC_DEPRECATED 408: 409: 2006-10-21 Havoc Pennington <hp@redhat.com> 410: 411: * Move all private API into the private Doxygen group sections of 412: the files, so they don't show up in the public API docs 413: 414: 2006-10-21 Havoc Pennington <hp@redhat.com> 415: 416: * dbus/dbus-message.h: put #ifndef DBUS_DISABLE_DEPRECATED around 417: dbus_message_iter_get_array_len(). 418: 419: * throughout: documentation improvements. 420: 421: 2006-10-20 Havoc Pennington <hp@redhat.com> 422: 423: * doc/TODO: remove the int64 thing from 1.0 since it doesn't 424: matter, and the message-loader-breaker thing since nobody is going 425: to do it. Add an item to 1.0 about supporting recursive locks 426: in dbus_threads_init_default() though, since it should be easy. 427: 428: * dbus/dbus-connection.c (_dbus_connection_read_write_dispatch): 429: Fix this in the !dispatch case to avoid busy-looping after 430: disconnection 431: 432: * More misc docs improvements 433: 434: 2006-10-19 Havoc Pennington <hp@redhat.com> 435: 436: * more Doxygen-related fixes (lots of moving things from the 437: public to internal section in dbus-connection.c) 438: 439: 2006-10-19 Havoc Pennington <hp@redhat.com> 440: 441: * Fix a pile of Doxygen warnings and missing docs 442: 443: 2006-10-19 John (J5) Palmieri <johnp@redhat.com> 444: 445: * bus/dir-watch-default.c, bus/dir-watch-dnotify.c, 446: bus/dir-watch-kqueue.c (bus_watch_directory): Pass in a BusContext 447: instead of a void *. kqueue uses this to get the context's loop 448: while the other modules ignore the parameter. This allows us to 449: avoid platform conditionals 450: 451: * bus/bus.c (process_config_postinit): Pass in the context to the 452: watch 453: 454: 2006-10-19 John (J5) Palmieri <johnp@redhat.com> 455: 456: * bus/messagebus.in, bus/rc.messagebus.in: run dbus-uuidgen --ensure 457: when starting the system bus 458: 459: 2006-10-18 John (J5) Palmieri <johnp@redhat.com> 460: 461: * configure.in: More fixups for cross compile (Patch from 462: Marco Pracucci <development at pracucci dot com>) 463: 464: 2006-10-17 Havoc Pennington <hp@redhat.com> 465: 466: * tools/dbus-send.c (main): don't close shared connection, fix 467: from Timo Hoenig 468: 469: 2006-10-17 Havoc Pennington <hp@redhat.com> 470: 471: * configure.in (CPPFLAGS): don't do -DDBUS_API_SUBJECT_TO_CHANGE here 472: 473: * dbus/dbus.h: drop the DBUS_API_SUBJECT_TO_CHANGE requirement, 474: since realistically it doesn't look like we'll be changing it 475: anymore. 476: 477: 2006-10-17 Havoc Pennington <hp@redhat.com> 478: 479: * dbus/dbus-internals.c (_dbus_warn_check_failed): new function to 480: be used for return_if_fail type warnings; prefixes the pid, and 481: fatal by default. 482: 483: 2006-10-14 John (J5) Palmieri <johnp@redhat.com> 484: 485: * configure.in: Released 1.0 RC 2(0.94) 486: Add check for -fPIC and enable it if available 487: 488: 2006-10-14 John (J5) Palmieri <johnp@redhat.com> 489: 490: * doc/TODO: Remove the check TODO item 491: 492: 2006-10-13 John (J5) Palmieri <johnp@redhat.com> 493: 494: * dbus-transport-socket.c (exchange_credentials): 495: Print out more detailed errors if reading or sending 496: credentials fail (Patch from Julio M. Merino Vidal 497: <jmmv at NetBSD dot org>) 498: 499: 2006-10-13 John (J5) Palmieri <johnp@redhat.com> 500: 501: * configure.in: when checking for posix getpwnam_r assume true 502: for cross compiles 503: 504: 2006-10-13 John (J5) Palmieri <johnp@redhat.com> 505: 506: * configure.in: Check for gethostbyname first before we check for it 507: in libnsl. On gnu systems it is implemeneted 508: 509: 2006-10-13 Havoc Pennington <hp@redhat.com> 510: 511: * dbus/dbus-connection.c 512: (dbus_connection_send_with_reply_and_block): fix to handle closed 513: connections, from Tambet Ingo bug #8631 514: 515: 2006-10-11 John (J5) Palmieri <johnp@redhat.com> 516: 517: * configure.in: use AC_TRY_COMPILE for dirfd instead of AC_TRY_RUN 518: 519: 2006-10-11 John (J5) Palmieri <johnp@redhat.com> 520: 521: * test/name-test/Makefile.am: don't link against both libdbus and 522: libdbus-convenience 523: 524: 2006-10-11 John (J5) Palmieri <johnp@redhat.com> 525: 526: * configure.in: check for dirfd even if it is a macro 527: (Patch from Timothy Redaelli <drizzt at gufi dot org>) 528: 529: 2006-10-08 John (J5) Palmieri <johnp@redhat.com> 530: 531: * configure.in: define _POSIX_PTHREAD_SEMANTICS on solaris 532: avoid using dirfd on systems that don't have dirfd 533: (Patch by Brian Cameron <brian dot cameron at sun dot com>) 534: 535: 2006-10-02 John (J5) Palmieir <johnp@redhat.com> 536: 537: * dbus/dbus-sysdeps.c (_dbus_abort): Remove from 538: #ifndef DBUS_DISABLE_ASSERTS macro to fix distcheck 539: 540: * dbus/dbus-sysdeps-unix.c (_dbus_print_backtrace): Remove from 541: #if !defined (DBUS_DISABLE_ASSERT) || defined(DBUS_BUILD_TESTS) 542: macro because _dbus_abort calls it 543: 544: * tools/Makefile.am: Add dbus-launch.h to the source list so distcheck 545: works 546: 547: 2006-10-02 John (J5) Palmieir <johnp@redhat.com> 548: 549: * dbus/dbus-sysdeps-util-unix.c (dirent_buf_size): Add check for 550: MAXNAMELEN for Mac OSX. (Patch from Jonas B <sanoix at gmail dot com>) 551: 552: 2006-10-02 Thiago Macieira <thiago@kde.org> 553: 554: * tools/dbus-launch.c (main): make it complain and abort 555: execution in case an unknown option is given. Also, process 556: the '--' argument as the end of arguments. 557: 558: 2006-10-01 Havoc Pennington <hp@redhat.com> 559: 560: * tools/dbus-launch.c (print_variables): if no syntax is given, 561: don't print something that's sort-of-half-sh-syntax, just print 562: a plain key-value pairs thing. 563: 564: * tools/dbus-launch-x11.c: use machine ID rather than hostname for 565: the local machine representation (but still have the hostname in 566: the display). Remove the hostname from the display if it is 567: localhost. Change session files to be named 568: ~/.dbus/session-bus/machine-display. Change X atoms to be 569: underscore-prefixed so nobody whines about ICCCM compliance. 570: Otherwise name them the same as the env variables. 571: Change session file format to include key-value pairs and an 572: explanatory comment. Keys are the same as the env variables. 573: (set_address_in_x11): X property format can't depend on 574: sizeof(pid_t) on a particular machine, fix to always be 32 bits 575: 576: * tools/dbus-launch.c: make --autolaunch take a machine id 577: argument. If --autolaunch is used with a program to run, complain 578: for now (but add a FIXME). Also, don't look for existing bus if 579: there's a program to run (but add a FIXME). 580: 581: * dbus/dbus-sysdeps-unix.c (_dbus_get_autolaunch_address): pass 582: machine uuid to dbus-launch (avoids linking dbus-launch to libdbus 583: just to get this, and avoids duplicating uuid-reading code). 584: 585: * tools/dbus-launch.1: clarify various things 586: 587: 2006-10-01 Havoc Pennington <hp@redhat.com> 588: 589: * test/test-service.c (path_message_func): remove broken extra 590: unref that was hidden by the bugs in dbus-connection.c/dbus-bus.c 591: 592: * test/test-shell-service.c (path_message_func): same fix 593: 594: * dbus/dbus-connection.c 595: (_dbus_connection_get_dispatch_status_unlocked): break up the 596: function a little for clarity and fix the notification of 597: dbus-bus.c to not require dispatch to be complete 598: 599: * dbus/dbus-connection.c (dbus_connection_unref): improve the 600: warning when you try to finalize an open connection. 601: 602: 2006-10-01 Havoc Pennington <hp@redhat.com> 603: 604: * dbus/dbus-bus.c 605: (internal_bus_get): only weak ref the connection; this means 606: _dbus_bus_notify_shared_connection_disconnected_unlocked can be 607: called safely in any context 608: (_dbus_bus_notify_shared_connection_disconnected_unlocked): don't 609: unref 610: 611: * dbus/dbus-connection.c 612: (_dbus_connection_get_dispatch_status_unlocked): move 613: _dbus_bus_notify_shared_connection_disconnected_unlocked here 614: when queuing Disconnected instead of when the Disconnected message 615: arrives, so dbus_bus_get() won't return closed connections. 616: 617: 2006-10-01 Havoc Pennington <hp@redhat.com> 618: 619: * dbus/dbus-connection.c (_dbus_connection_close_if_only_one_ref): 620: Add a hack to make DBusNewConnectionFunction work right. 621: 622: * dbus/dbus-server-socket.c (handle_new_client_fd_and_unlock): use 623: the hack here. Also, fix the todo about refcount leak. 624: 625: * dbus/dbus-server-debug-pipe.c (_dbus_transport_debug_pipe_new): 626: and use the hack here 627: 628: * dbus/dbus-connection.c: Kill the "shared" flag vs. the 629: "shareable" flag; this was completely broken, since it meant 630: dbus_connection_open() returned a connection of unknown 631: shared-ness. Now, we always hold a ref on anything opened 632: as shareable. 633: 634: Move the call to notify dbus-bus.c into 635: connection_forget_shared_unlocked, so libdbus consistently forgets 636: all its knowledge of a connection at once. This exposed numerous 637: places where things were totally broken if we dropped a ref inside 638: get_dispatch_status_unlocked where 639: connection_forget_shared_unlocked was previously, so move 640: connection_forget_shared_unlocked into 641: _dbus_connection_update_dispatch_status_and_unlock. Also move the 642: exit_on_disconnect here. 643: 644: (shared_connections_shutdown): this assumed weak refs to the 645: shared connections; since we have strong refs now, the assertion 646: was failing and stuff was left in the hash. Fix it to close 647: still-open shared connections. 648: 649: * bus/dispatch.c: fixup to use dbus_connection_open_private on the 650: debug pipe connections 651: 652: * dbus/dbus-connection.c (dbus_connection_dispatch): only notify 653: dbus-bus.c if the closed connection is in fact shared 654: (_dbus_connection_close_possibly_shared): rename from 655: _dbus_connection_close_internal 656: (dbus_connection_close, dbus_connection_open, 657: dbus_connection_open_private): Improve docs to explain the deal 658: with when you should close or unref or both 659: 660: * dbus/dbus-bus.c 661: (_dbus_bus_notify_shared_connection_disconnected_unlocked): rename 662: from _dbus_bus_check_connection_and_unref_unlocked and modify to 663: loop over all connections 664: 665: * test/test-utils.c (test_connection_shutdown): don't try to close 666: shared connections. 667: 668: * test/name-test/test-threads-init.c (main): fix warnings in here 669: 670: * dbus/dbus-sysdeps.c (_dbus_abort): support DBUS_BLOCK_ON_ABORT 671: env variable to cause blocking waiting for gdb; drop 672: DBUS_PRINT_BACKTRACE and just call _dbus_print_backtrace() 673: unconditionally. 674: 675: * configure.in: add -export-dynamic to libtool flags if assertions enabled 676: so _dbus_print_backtrace works. 677: 678: * dbus/dbus-sysdeps-unix.c (_dbus_print_backtrace): use fprintf 679: instead of _dbus_verbose to print the backtrace, and diagnose lack 680: of -rdynamic/-export-dynamic 681: 682: 2006-09-30 Havoc Pennington <hp@redhat.com> 683: 684: * dbus/dbus-bus.c (dbus_bus_get_private, dbus_bus_get) 685: (internal_bus_get): fix screwy code formatting. whoever committed 686: that was not paying attention! 687: 688: 2006-09-30 Havoc Pennington <hp@redhat.com> 689: 690: * configure.in (LT_CURRENT, LT_AGE): increment current and age to 691: reflect addition of interfaces. 692: 693: * doc/dbus-specification.xml: describe a new 694: org.freedesktop.DBus.Peer.GetMachineId method 695: 696: * dbus/dbus-string.c (_dbus_string_skip_white_reverse): new function 697: (_dbus_string_skip_white, _dbus_string_skip_blank): use new 698: DBUS_IS_ASCII_BLANK, DBUS_IS_ASCII_WHITE macros and fix assertion 699: at end of skip_white 700: (_dbus_string_chop_white): new function 701: 702: * bus/connection.c (bus_connections_setup_connection): call 703: dbus_connection_set_route_peer_messages. 704: 705: * dbus/dbus-connection.c 706: (_dbus_connection_peer_filter_unlocked_no_update): modify to 707: support a GetMachineId method. 708: 709: Also, support a new flag to let the bus pass peer methods through 710: to apps on the bus, which can be set with 711: dbus_connection_set_route_peer_messages. 712: 713: Finally, handle and return an error for anything unknown on the 714: Peer interface, which will allow us to extend the Peer interface 715: in the future without fear that we're now intercepting something 716: apps were wanting to see. 717: 718: * tools/dbus-uuidgen.c: a thin wrapper around the functions in 719: dbus/dbus-uuidgen.c 720: 721: * dbus/dbus-uuidgen.c: implement the bulk of the dbus-uuidgen 722: binary here, since most of the code is already in libdbus 723: 724: * dbus/dbus-sysdeps.c (_dbus_read_local_machine_uuid): read the 725: uuid from the system config file 726: 727: * dbus/dbus-internals.c (_dbus_generate_uuid, _dbus_uuid_encode) 728: (_dbus_read_uuid_file_without_creating) 729: (_dbus_create_uuid_file_exclusively, _dbus_read_uuid_file): new 730: uuid-related functions, partly factored out from dbus-server.c 731: 732: * dbus/dbus-sysdeps.c (_dbus_error_from_errno): convert EEXIST to 733: DBUS_ERROR_FILE_EXISTS instead of EEXIST 734: 735: * dbus/dbus-protocol.h (DBUS_ERROR_FILE_EXISTS): add file exists error 736: 737: * tools/dbus-cleanup-sockets.1: explain what the point of this 738: thing is a bit more 739: 740: * autogen.sh (run_configure): add --config-cache to default 741: configure args 742: 743: * dbus/dbus-internals.h (_DBUS_ASSERT_ERROR_IS_SET): disable the 744: error set/clear assertions when DBUS_DISABLE_CHECKS is defined 745: 746: * tools/dbus-launch.c (main): if xdisplay hasn't been opened, 747: don't try to save address, fixes crash in make check 748: 749: 2006-09-30 Thiago Macieira <thiago@kde.org> 750: 751: * configure.in: add DBUS_BINDIR as a #define to C source code. 752: 753: * tools/dbus-launch.c 754: * tools/dbus-launch.h 755: * tools/dbus-launch-x11.c: 756: * tools/dbus-launch.1: Add the --autolaunch option to 757: dbus-launch, which makes it scan for an existing session 758: started with --autolaunch. With that option, it also creates 759: an X11 window and saves the bus address and PID to it. 760: 761: * dbus/dbus-sysdeps.h: 762: * dbus/dbus-sysdeps-unix.c (_dbus_get_autolaunch_address): Add 763: a function that runs "dbus-launch --autolaunch" to retrieve 764: the running D-Bus session address (or start one if none was running) 765: 766: * dbus/dbus-transport.c: Add the handler for the "autolaunch:" 767: address protocol, which tries to get the running session from 768: dbus-launch. 769: 770: * dbus/dbus-bus.c: 771: * dbus/dbus-internals.h: Make "autolaunch:" be the default 772: D-Bus session bus address. 773: 774: * dbus/dbus-connection.c: Fix horrible typo in error message. 775: 776: 2006-09-18 John (J5) Palmieri <johnp@redhat.com> 777: 778: * tools/Makefile.am: use @EXPANDED_DATADIR@ instead of @DATADIRNAME@ 779: 780: 2006-09-17 Havoc Pennington <hp@redhat.com> 781: 782: * dbus/dbus-transport.c (_dbus_transport_get_is_authenticated): 783: fix so that if unix_user_function is set on windows, we still 784: do the default check that the auth identity matches the bus identity 785: