summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Fix some transient window issueswip/mir3.10Robert Ancell2014-06-111-14/+51
|
* Add some more debugging infoRobert Ancell2014-06-112-82/+89
|
* Don't create surfaces for input only windowsRobert Ancell2014-06-111-0/+2
|
* Don't crash on unknown propertiesRobert Ancell2014-06-101-1/+7
|
* Move debugging into a separate fileRobert Ancell2014-06-104-276/+307
|
* Regenerate surface on resizeRobert Ancell2014-06-101-2/+12
|
* Make GdkMirWindowImpl private againRobert Ancell2014-06-094-73/+105
|
* Unref window when pointer object is finalizedRobert Ancell2014-06-091-0/+14
|
* Draw transient windows onto their parentsRobert Ancell2014-05-296-31/+145
|
* Whoops, I forgot the magic 8Robert Ancell2014-05-291-1/+1
|
* Provide gtk-error-bell settingRobert Ancell2014-05-291-0/+6
|
* mir: more gtksettings...Ryan Lortie2014-05-291-0/+18
|
* mir: handle resize eventsRyan Lortie2014-05-292-3/+46
| | | | | ...including the possibility that we don't receive the buffer size that we request.
* build: add mir .pc files to distcleanRyan Lortie2014-05-291-0/+2
|
* mir: one more gtksetting...Ryan Lortie2014-05-281-0/+6
|
* mir: set core pointer on GdkDisplayRyan Lortie2014-05-281-0/+3
|
* mir: one more gtksetting...Ryan Lortie2014-05-281-0/+6
|
* mir: allow windows to become focusedRyan Lortie2014-05-282-4/+36
| | | | | | | | Send focus change events to toplevels in response to getting the event from mir. Now that the window is focused, cursor blink works, so wire up those GtkSettings as well.
* mir: fill in some more GtkSettingsRyan Lortie2014-05-281-0/+98
| | | | ...and add an assert for those we don't handle.
* Correct spelling of Ambiance theme (still doesn't work though)Robert Ancell2014-05-281-1/+1
|
* mir: we must leak the window ref for nowRyan Lortie2014-05-282-8/+5
| | | | | | By the time we get the call to ensure_no_surface() the window is already being disposed and the qdata is gone. Our attempt to free the copy of the ref owned by Mir by looking it up again from our qdata will fail.
* Set KeyEvent string fieldRobert Ancell2014-05-281-3/+60
|
* mir: do proper event dispatchingRyan Lortie2014-05-285-601/+871
| | | | | | | | | | | | | | We had a race where destroying a window while an event is being dispatched to it could result in a crash. We can't just hold a ref on the window because the object is externally visible and we could end up running a binding's toggle ref handler in the mir thread. Add a proper source type with a queue and try to do this properly by dispatching the event back to the main thread before we check if the window still exists (through use of a weak ref). This commit also moves all of the event translation code out of gdkmirwindowimpl.c (which is getting a bit big...).
* Counter suspicious hover-exit eventsRobert Ancell2014-05-281-2/+13
|
* Generate crossing eventsRobert Ancell2014-05-281-9/+34
|
* Let GDK decide if it wants events, ignore the event maskRobert Ancell2014-05-272-24/+15
|
* Remove debuggingRobert Ancell2014-05-271-6/+0
|
* Correctly hide windowsRobert Ancell2014-05-276-59/+92
|
* Handle Mir events on the main threadRobert Ancell2014-05-271-85/+137
|
* Finalize MirWindowImpl valuesRobert Ancell2014-05-271-0/+9
|
* Implement gdk_mir_window_impl_get_device_stateRobert Ancell2014-05-271-2/+7
|
* Get menus working (sort of)Robert Ancell2014-05-271-6/+7
|
* Fix return windows in gdk_mir_pointer_query_stateRobert Ancell2014-05-271-2/+4
|
* Use event queue (sort of) correctlyRobert Ancell2014-05-272-4/+13
|
* Disable debugging messageRobert Ancell2014-05-271-1/+1
|
* Disable debugging messageRobert Ancell2014-05-271-1/+1
|
* Flesh out keymapRobert Ancell2014-05-271-30/+322
|
* Emit eventsRobert Ancell2014-05-271-2/+3
|
* Don't set send_event, make public function privateRobert Ancell2014-05-271-2/+1
|
* Link associated devicesRobert Ancell2014-05-271-0/+3
|
* Make separate keyboard and pointer device classesRobert Ancell2014-05-279-195/+471
|
* Set initial window event mask, don't set event variables we don't needRobert Ancell2014-05-264-16/+9
|
* Set more event fieldsRobert Ancell2014-05-261-4/+5
|
* mir: create devices from _constructedRyan Lortie2014-05-261-0/+11
| | | | The "display" property is not yet set at time of _init.
* Stub Mir moduleRobert Ancell2014-05-2620-6/+3383
|
* Bump versionMatthias Clasen2014-05-121-2/+2
|
* 3.10.93.10.9Matthias Clasen2014-05-121-0/+15
|
* a11y: Avoid a warningMatthias Clasen2014-05-121-1/+2
| | | | | The listbox row accessible code was giving warnings when parent is NULL, at destruction time.
* applicationwindow: Properly unmap the menubar widgetBenjamin Otte2014-05-121-0/+13
|
* GtkColorChooserWidget: Redraw swatches when selection changesMarcus Karlsson2014-05-121-1/+6
| | | | | | | | | | | When selecting a swatch in a GtkColorChooserWidget the previously selected swatch and the currently selected swatch are not redrawn. This can leave the old swatch still marked with a checkbox even though a new swatch has been selected. Redraw the swatches after changing the selection. https://bugzilla.gnome.org/show_bug.cgi?id=727487
View Lorry Controller Status