aboutsummaryrefslogtreecommitdiff
path: root/libdbusmenu-glib
Commit message (Collapse)AuthorAgeFilesLines
* Fix build with --enable-massivedebuggingDmitry Shachnev2016-11-241-1/+1
| | | | Use %p for printing pointers, otherwise GCC complains.
* Use gi's typelibdir and install in its (multiarch) locationsIain Lane2014-10-301-1/+1
|
* add gtk-doc markup to our subclasses' structs 'parents' fieldsCharles Kerr2014-03-183-0/+3
|
* Protect client_new functions from bad dbus names and objects.Ted Gould2013-09-091-0/+7
|\ | | | | | | Approved by PS Jenkins bot, Charles Kerr.
| * Throw criticals on bad names and/or objectsTed Gould2013-06-101-0/+7
| |
* | Fix breakage when building with coverage scanning.Charles Kerr2013-06-201-3/+0
|\ \ | |/ |/| | | Approved by PS Jenkins bot, Ted Gould.
| * fix an error introduced in r399 that broke g-ir-scanner invocation when ↵Charles Kerr2012-05-021-3/+0
| | | | | | | | configure is invoked with --enable-gcov
* | Update for review commentsChris Coulson2013-05-281-5/+11
| |
* | Set the max ID of menuitems to G_MAXINT rather than 30000, as internally, ↵Chris Coulson2013-05-171-2/+5
| | | | | | | | | | | | the server side happily creates ID's greater than this which then wraparound on the client side. We keep the minimum unchanged at -1, as some consumers addume that < 0 is an error (rather than -1). To make this a bit less fragile, handle rollover on the server side from G_MAXINT to 1, rather than having a signed integer overflow which will just break things a bit later on
* | Fix multiple leaks due to improper use of g_variant_parse()Chris Coulson2013-01-242-5/+14
| |
* | Make sure we unref the GDBusMethodInvocation when not sending a replyChris Coulson2013-01-221-0/+6
| |
* | gtk-doc fix: dbusmenu_menuitem_set_root doesn't have a return valueCharles Kerr2012-12-041-2/+0
| |
* | gtk-doc fix: DbusmenuMenuitemClass.send_about_to_showCharles Kerr2012-12-041-1/+1
| |
* | when initializing fields inside a struct being declared, replace gccism with ↵Charles Kerr2012-10-262-5/+5
|\ \ | | | | | | | | | | | | | | | c99 syntax. Approved by PS Jenkins bot, Albert Astals Cid.
| * | when initializing fields inside a struct being declared, replace gccism with ↵Charles Kerr2012-10-262-5/+5
| | | | | | | | | | | | c99 syntax
* | | null type means standard typeAlbert Astals2012-10-251-1/+1
| | |
* | | Use g_variant_compareAlbert Astals2012-10-251-1/+1
| | |
* | | Do not reuse menu items if their type is differentAlbert Astals2012-10-251-2/+23
|/ / | | | | | | | | | | | | The gtk side does not know how to convert from a separator to a standard item, and even if it'd knew it does not make much sense to reuse them since if the types are different you can expect there's little to reuse
* | fix grammar: use of its and it'sCharles Kerr2012-10-075-10/+10
| |
* | revert r417 so that we don't change the vapi name at the last minute for ↵Charles Kerr2012-09-201-3/+3
| | | | | | | | 12.10 (LP: #978687)
* | merge lp:~charlesk/dbusmenu/constify-dbusmenu-menuitem-gettersCharles Kerr2012-08-212-13/+13
|\ \
| * | constify the dbusmenu_menuitem_get*() functionsCharles Kerr2012-04-252-13/+13
| | |
* | | don't include build-time filenames in comments of the enum header filesunknown2012-05-241-1/+1
| |/ |/| | | | | | | | | | | | | | | If we do, the header files may be different on each build. This is especially harmful when building multiple times for multiple architectures and expecting the files to be identical. See: http://bugs.debian.org/674200
* | Evan Nemerson's patch to make the Vala bindings use pkg-config names, rather ↵Charles Kerr2012-04-261-3/+3
|\ \ | |/ |/| | | than GIR names, so that valac can automatically pick up the relevant libraries and flags.
| * apply Evan Nemerson's patch to make the Vala binidngs use pkg-config names, ↵Charles Kerr2012-04-111-3/+3
| | | | | | | | not GIR names, so that valac can automatically pick up the relevant libraries and flags.
* | fix leaky DbusmenuServer.priv.dbusobjectCharles Kerr2012-04-141-0/+5
|/
* In menuitem_signals_remove(), disconnect the ↵Charles Kerr2012-04-101-0/+1
| | | | | | | DBUSMENU_MENUITEM_SIGNAL_SHOW_TO_USER event that we connected to in menuitem_signals_create(). It looks like this non-disconnected signal goes all the way back to July 2010 <http://bazaar.launchpad.net/~dbusmenu-team/dbusmenu/trunk.0.6/revision/139#libdbusmenu-glib/server.c> so I'd like ted's eyes on this to confirm there's not some nuance that I'm missing...
* merge lp:~ted/dbusmenu/about-to-show-tuple to fix about-to-show grouping.Charles Kerr2012-04-091-3/+5
|\
| * Apparently this type is wrong, if only this language supported some way to ↵Ted Gould2012-04-091-1/+1
| | | | | | | | communicate to programmers the types that variables were. Oh, we can dream.
| * Pull the items out of the tupleTed Gould2012-04-091-2/+4
| |
* | make event_data_find() and session_bus_cb() private instead of publicCharles Kerr2012-04-051-2/+2
| |
* | Make sure that if the callback is NULL we're not trying to get a callbackTed Gould2012-04-041-1/+5
| |
* | Protect from the cases where we might get NULL in the data segmentTed Gould2012-04-041-1/+8
|/
* Merge lp:~ted/dbusmenu/event-grouping to reduce the number of dbus messages ↵Charles Kerr2012-04-044-54/+666
|\ | | | | | | | | | | | | that hud-service sends. This change takes all of the events on a client and puts them into single dbus messages to be sent to the server. It also does the same thing for about-to-show messages. Both sets of messages are reserialized on either side to make callers unable to notice the difference.
| * Fixing a cut-and-paste commentTed Gould2012-04-031-1/+1
| |
| * Switching to GQueueTed Gould2012-04-031-28/+36
| |
| * Fill out the group callback to update if we need itTed Gould2012-04-031-0/+30
| |
| * Fleshing out the idle handler for about to showTed Gould2012-04-031-0/+74
| |
| * Setup our idle if we're grouping the eventsTed Gould2012-04-031-8/+37
| |
| * Set up the about-to-show task tracking variablesTed Gould2012-04-031-1/+28
| |
| * Pulling out parts of about to show so that we have a core function thereTed Gould2012-04-031-12/+21
| |
| * Set and get the event grouping and notify if we change itTed Gould2012-04-031-0/+13
| |
| * Add a property to whether we should group events or notTed Gould2012-04-032-1/+12
| |
| * Making sure to clear the idle so we queue againTed Gould2012-03-301-0/+1
| |
| * Unpacking the tupleTed Gould2012-03-301-1/+3
| |
| * Making sure to init our errorTed Gould2012-03-301-1/+1
| |
| * Making sure the variant is appropriately wrappedTed Gould2012-03-301-2/+3
| |
| * If we're going to unref we should set to NULL to catch potential errorsTed Gould2012-03-301-0/+3
| |
| * Checking the version of the server to see if we can group the propertiesTed Gould2012-03-301-0/+32
| |
| * Okay, making all the callbacks real and returning appropriate errorsTed Gould2012-03-301-1/+53
| |
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-21 10:44:55 +0000