summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Set up attributes properly for wl_shell transient windowsdevs/derekf/apbbbugsDerek Foreman2016-06-291-0/+14
| | | | This fixes Qt sub-windows.
* Only unignore a client on the first commit with a bufferDerek Foreman2016-06-291-1/+1
| | | | | Otherwise, if a wayland client does its first commit without a buffer attached it will confuse our focus logic.
* destroy evry's focus-out timer if focus-in occurs before timer triggersMike Blumenkrantz2016-06-291-0/+11
| | | | fixes dismissal during comp input regrabs
* Add a pixel get callback for mirrorsDerek Foreman2016-06-281-6/+21
| | | | | | | | | | Mirrors can be rendered independently of what they're mirroring, which (at least under wayland) can result in a situation where the mirror is rendered before the parent sets up their image pointers properly. We give mirrors their own callback to prevent that from causing a crash.
* Revert "Always clear the mirror image data pointer"Derek Foreman2016-06-281-1/+2
| | | | | | | This reverts commit 326d638ba274ba4623c4112a68250b47f7b250e1. This, for some reason, resulted in client surfaces being blacked out, despite only ever changing mirror surfaces.
* Fix wayland pixmap ids againDerek Foreman2016-06-281-2/+10
| | | | | | | | Wayland pixmap ids are a different data type for internal and external windows. cast them both to 64-bits so they're the same size regardless of arch. ref d3ba524a62713ff58c1537524030c155d0713ca2
* emix: update sink of sink_input when changedMarcel Hollerbach2016-06-251-0/+7
| | | | this fixes none changing sinks in the app.
* e_comp: silence warning, this time reallyMarcel Hollerbach2016-06-251-1/+1
| | | | as said on the ML
* e_comp: silence warningMarcel Hollerbach2016-06-251-1/+1
| | | | after talking to zmike, this is by intend.
* e_xkb: eval group before setted to wlMarcel Hollerbach2016-06-251-14/+30
| | | | | | | otherwise we are going to set 0 as default group. This commit refactors out _eval_cur_group from the init timer, since wl does not need a init timer.
* music-player: add spotifyMarcel Hollerbach2016-06-251-0/+1
|
* Fix zone setup for wayland compositor on XDerek Foreman2016-06-241-1/+1
| | | | | | Both the X and WL code try to set up the output pointer for the zone. Make the WL output take precedence, since it's actually used for events.
* add some hacks to route mouse events to wl popups correctly while mouse is downMike Blumenkrantz2016-06-241-1/+49
| | | | | | | | in a choice between fixing a corner case popup behavior and breaking dnd or having functional dnd and adding hacks to fix corner case popup behavior, adding more hacks was the obvious correct solution ref 03a4ecbdb0972f8267c07ea14d752fcee69bd9fa
* remove unused event info from _e_comp_wl_mouse_out()Mike Blumenkrantz2016-06-241-5/+5
|
* Don't try to send surface presence events for zones with no wayland outputDerek Foreman2016-06-241-0/+2
| | | | Avoid that NULL pointer deref.
* Revert "set NOGRAB pointer mode on comp object event objects"Mike Blumenkrantz2016-06-231-2/+0
| | | | | | This reverts commit 03a4ecbdb0972f8267c07ea14d752fcee69bd9fa. this should be correct, and yet it crashes xwayland and breaks dnd
* set NOGRAB pointer mode on comp object event objectsMike Blumenkrantz2016-06-231-0/+2
| | | | fixes dragging off surfaces in wayland
* Don't send configure to wl_shell popupsDerek Foreman2016-06-231-0/+11
| | | | | This confuses Qt badly, and rightly so, because we shouldn't be doing it.
* Use e_client_util_is_popup for checking wayland clientsDerek Foreman2016-06-231-9/+5
| | | | | All the different popup types are tested by the helper function, the current wayland tests were missing some.
* disallow pager window drags for popups and ignored windowsMike Blumenkrantz2016-06-232-0/+2
|
* Revert "do not change E_Client->saved values during maximize if ↵Mike Blumenkrantz2016-06-221-20/+4
| | | | | | | | need_maximize is set" This reverts commit 9a9c09f514d02edf7feb6770cefacf0b3f77f2be. a better solution should be found at some point since this is broken in x11 too
* don't set need_maximize during wl commit maximizeMike Blumenkrantz2016-06-221-2/+0
| | | | | | this breaks the maximize! partially reverts b5576dbb8c98934ab0e8c5ff59bed106c02d9874
* freeze events on comp object input rects during deletionMike Blumenkrantz2016-06-221-0/+1
| | | | prevents some event propagation which would cause flickering
* Always clear the mirror image data pointerDerek Foreman2016-06-211-2/+1
| | | | | | | Under wayland evas will sometimes use the old one, I have no idea why. Fixes a crash bug when mousing out of menus in a GTK app under wayland. fix T3576
* e_xkb: init the keyboard on initMarcel Hollerbach2016-06-211-1/+2
| | | | | we need to flush in the first group here. For wayland currently each keyboard configuration needs to be flushed each time.
* Update some wl_shell window handling logicDerek Foreman2016-06-211-1/+16
| | | | | | | | | Our xdg_shell implementation has received some fixes that are also applicable to wl_shell. Qt still uses wl_shell, so we do have a test case, and do need to keep it up to date. This copies some window attributes, clamping logic, and diagnostic prints from the xdg_shell implementation.
* Send wayland surface enter and leave eventsDerek Foreman2016-06-212-0/+54
| | | | | These events are supposed to be sent to clients that have bound wl_output and manipulate surfaces.
* Track X outputs for zonesDerek Foreman2016-06-211-0/+3
| | | | | We needed this for wayland, so we add the same tracking for X to be consistent - though we have no users yet.
* Track wayland outputs for zonesDerek Foreman2016-06-212-0/+7
|
* Add new API for finding a zone by its id stringDerek Foreman2016-06-212-0/+19
|
* apply client size for session recovery rememberMike Blumenkrantz2016-06-201-1/+1
| | | | fixes maximize restoring
* use separate flag for pending wl client unmaximizeMike Blumenkrantz2016-06-202-3/+12
| | | | also set flag to prevent saved geometry from being overwritten
* do not change E_Client->saved values during maximize if need_maximize is setMike Blumenkrantz2016-06-201-4/+20
| | | | | this is indicative of a deferred maximize, so don't re-set it and screw up previously-set values
* store maximize property from E_Remember to diskMike Blumenkrantz2016-06-201-0/+1
|
* set need_maximize flag on remembered client when applying maximizeMike Blumenkrantz2016-06-201-0/+1
| | | | causes maximize to be applied
* use pending wl maximize value to determine maximize flag in xdg configureMike Blumenkrantz2016-06-201-3/+3
| | | | fixes some cases where fast configures would break maximizing
* match deskmirror client opacity with original objectMike Blumenkrantz2016-06-201-0/+13
| | | | fix T3911
* update secondary remembers when a session recovery remember is updatedMike Blumenkrantz2016-06-201-0/+5
| | | | | the session recovery remember may block updates for the user-created remember, creating continuity issues when a new instance is launched
* prevent E_Remember->used_count from going negativeMike Blumenkrantz2016-06-201-1/+1
|
* prevent editing of session recovery remembersMike Blumenkrantz2016-06-201-10/+18
| | | | find/create new ones in this case
* replace wireless connection error #warning with notificationMike Blumenkrantz2016-06-201-1/+16
| | | | | this needs a more robust infrastructure than I'm prepared to write at the moment, but errors do need to be seen
* enforce mouse grabs and block client/compositor event propagation while activeMike Blumenkrantz2016-06-182-4/+8
| | | | | fixes cases where mouse events could filter to objects and cause unwanted effects, such as evry hiding randomly under wayland
* Fix eo object use after deletion in evryDerek Foreman2016-06-171-0/+3
| | | | | | | On window hide the focus out callback is triggering and causing a new timer to be set up. This timer fires when the window no longer exists. Remove the focus_out callback when we hide the window.
* force unset override and do reshadow on evry when switching to normal win modeMike Blumenkrantz2016-06-171-0/+2
| | | | fixes window movement and close animations
* Fix wl_shell popup livelockDerek Foreman2016-06-161-1/+0
| | | | | | | | | Trying to put wl_shell popups in the popup layer results in stacking problems later and a never ending loop. Just don't. fix T3433
* do not mouse-out of clients on focus out when focusing a popupMike Blumenkrantz2016-06-161-1/+1
| | | | | | | popups are transient in nature and do not interfere with mouse-based focus policies ref 3e6d6b348fd856d161aab98f5024e2293c292bd9
* Fix wl_shell clamping rulesDerek Foreman2016-06-151-2/+2
| | | | | The current clamping logic clamps submenus so that they're contained within their parent instead of adjacent to them.
* Fix crash in wl_shell surface destroyDerek Foreman2016-06-151-2/+6
| | | | | wl_shell doesn't set up shell.data, but this destructor is shared with xdg_shell which does.
* add separate mouse in/out handlers for internal winsMike Blumenkrantz2016-06-132-14/+71
| | | | | | | | internal wayland windows are windows with ssd, meaning they can only receive pointer events on the contents of the window and not the entire window including decoration regions ref T3819
* use focused client for window actions if passed object is not a client objectMike Blumenkrantz2016-06-091-146/+54
| | | | | | | | in the case where an action is triggered from the compositor or manager contexts the passed object will not be a client, causing actions to fail when they should succeed fix T3854
View Lorry Controller Status