Commit graph

106 commits

Author SHA1 Message Date
Martin Gräßlin
24bdfc6317 Make Toplevel::window() no longer virtual and introduce ::windowId()
Summary:
Toplevel::window() is the actual X11 window. This makes it difficult
to use as the generic identifier for both X11 and Wayland. The Wayland
ShellClient already had a windowId() which is now added to Toplevel as
a virtual method. On X11 (Toplevel default) it returns the window().

The method window() now returns XCB_WINDOW_NONE for classes without
the Toplevel::m_client, such as ShellClient. Thus it allows to properly
check whether we are on Wayland or X11.

The code is adjusted to use windowId where a generic id is needed and
to properly check whether the window is valid before using it where
a window() is used.

This also fixes at least one additional unknown issue in
Workspace::setActiveClient

where the windowId of a Wayland client was passed to X11.

Reviewers: #plasma

Subscribers: plasma-devel

Projects: #plasma

Differential Revision: https://phabricator.kde.org/D1527
2016-05-17 14:13:24 +02:00
Martin Gräßlin
013de158b7 Move ShellClient initialization code into a dedicated init method
Summary:
Preparation for also supporting XdgShell. There will be different
ctors for ShellSurface and XdgShell, but most code needs to be shared.
Thus a dedicated init method is needed.

There is some restructuring in the init. All code depending on
ShellSurface being set is grouped and in an if block.

Reviewers: #plasma

Subscribers: plasma-devel

Projects: #plasma

Differential Revision: https://phabricator.kde.org/D1440
2016-04-19 13:28:54 +02:00
Martin Gräßlin
ba7c5bbd77 Add support for opacity in ShellClient 2016-02-18 08:27:02 +01:00
Martin Gräßlin
ecce036979 [wayland] Introduce a memeber variable for checking whether ShellClient is internal
ASAN complained about heap-use-after-freeze in copyToDeleted due to call
into isInternal().
2015-12-18 17:26:21 +01:00
Martin Gräßlin
5e96f65224 Add window decoration to ShellClient
If a ShellClient supports the ServerSideDecoration interface we can
create a server decoration for it. For that updateDecoration is added
as a pure virtual method in AbstractClient and a more-or-less code copy
from Client is added to ShellClient.

Geometry handling is adjusted to consider the window decoration offsets.
2015-12-18 16:41:49 +01:00
Martin Gräßlin
5cef26d275 [wayland] Add support for ServerSideDecorationManager
We announce support for it and depending on whether we have a plugin
or not set the default mode to Server or None.

When a decoration interface is created it gets installed on the
ShellClient. But there it isn't properly used yet as we don't have
support for decorations in ShellClient yet.
2015-12-18 16:41:49 +01:00
Martin Gräßlin
3de3a959c6 Move implementation of clientPos() to AbstractClient
The Client's variant works for both Client and ShellClient.
2015-12-18 16:41:49 +01:00
Martin Gräßlin
827486ff36 Introduce a Toplevel::clientContentPos() -> QPoint
This describes an additional offset for the client content. On X11
our client content position matches with the window - the window
decoration is part of the overall content coordinate system.

On Wayland the content is an own texture starting at 0/0. Thus a
mapping to texture coordinates will be required when server side
decorations are provided. The new information is used in the scene's
to adjust the rendering and generating of quads.
2015-12-18 16:41:49 +01:00
Martin Gräßlin
5f90fa5cfd Introduce pure virtual, protectd AbstractClient::acceptsFocus -> bool
Replacement for calls to info->input() which is only valid for the Client
sub class, but not for ShellClient.

In ShellClient the implementation is swapped with wantsInput() and
wantsInput() has a new implementation which properly delegates to rules()
just like Client does.
2015-12-18 16:41:49 +01:00
Martin Gräßlin
a4c347fc7a [wayland] Translate inputTransformation matrix by borderLeft/Top
With decorations the ShellClient needs to have another input offset
by borderTop and borderLeft.
2015-12-18 16:41:49 +01:00
Bhushan Shah
450bbaafdc [wayland] Introduce property to identify lockscreen and inputmethods
This introduces Toplevel::isLockScreen() and Toplevel::isInputMethod(),
this can be used to allow only lockscreen/inputmethods to get input
events and shown when screen is locked.
2015-11-16 15:49:38 +05:30
Bhushan Shah
3a1371989e [wayland] Introduce ShellClient::isLockScreen method
This allows to check if specific ShellClient is from LockScreen or not,
as well as this adds method to verify if ShellClient is from input
method like maliit.

Now that KWin knows about which window is from Screenlocker it can apply
various security restrictions like no other window then greeter is on
top of it.

Reviewed-By: Martin Gräßlin
2015-11-06 19:44:13 +05:30
Marco Martin
1f9fa64a49 preliminar support for task geometries in wayland
make the minimize effect work by reading taskGeometry
from plasmawindowmanagement and returning as iconGeometry()
there is one task geometry per panel window, iconGeometry()
will return the geometry associated to the nearest panel
from the window

REVIEW:125873
2015-11-05 11:36:40 +01:00
Martin Gräßlin
89a4df651d [wayland] Sync resizes
Similar to X11 world: we send a sync request on each size change and
block till we get the next damage with the proper size.

Testing seems to show a very smooth resize experience. We automatically
sync to the resize speed of the client.

Maybe we need a timeout in case the client isn't able to resize to the
requested size.
2015-10-28 11:16:48 +01:00
Martin Gräßlin
8cdfb0e88f Merge back implementation of Client::move and ShellClient::move
Implementation goes to AbstractClient, method is no longer virtual.
The X11 specific code is moved to a new virtual protected doMove
method implemented in Client.
2015-10-15 15:29:09 +02:00
Martin Gräßlin
6049b9ff3c Move implementation of sendToScreen from Client to AbstractClient
Method is no longer virtual, stub in ShellClient removed.
2015-10-14 16:38:17 +02:00
Martin Gräßlin
9f91431e8c Move quick tiling from Client to AbstractClient
In order to make it possible to quick tile also ShellClients.
2015-10-14 08:36:29 +02:00
Martin Gräßlin
15f5886426 Add a pure virtual AbstractClient::setGeometryRestore(const QRect &) 2015-10-14 08:36:29 +02:00
Martin Gräßlin
ec6c778df7 Move implementation of checkWorkspacePosition to AbstractClient
No longer a virtual method, the implementation of Client is turned into
AbstractClient. The stub removed from ShellClient.
2015-10-14 08:36:29 +02:00
Martin Gräßlin
696cdb9e39 Add setGeometry to AbstractClient as pure virtual method
In Client existing method is marked as override, in ShellClient a new
override is added which delegates to requestGeometry. Existing method
is renamed to doSetGeometry and all internal calls delegat to it.
Needs better merging with the implementation of Client.
2015-10-14 08:36:29 +02:00
Martin Gräßlin
eacaf25acf Improve sharing of maximize between Client and ShellClient
The changeMaximize method is added as a pure virtual protected method
to AbstractClient. This replaces the previous pure virtual maximize
method. Which is now directly implemented in AbstractClient (reusing
the implementation previously in Client).
2015-10-14 08:36:29 +02:00
Martin Gräßlin
32fbbc90bc Make geometryRestore() pure virtual in AbstractClient
Provides an implementation in ShellClient and method in Client gets
marked as override.
2015-10-12 10:23:06 +02:00
Martin Gräßlin
0bb4498e1d [wayland] Proper metatype for ShellClient
Needs meta type declared and properly used in WaylandServer.
2015-10-08 12:51:40 +02:00
Martin Gräßlin
3cb8438399 [wayland] Keep transient state directly in ShellClient
So far we delegated to ShellSurfaceInterface, but it's possible that
we call into isTransient when the ShellSurfaceInterface is already
destroyed.
2015-10-08 11:04:06 +02:00
Martin Gräßlin
b35c3e8501 [wayland] Keep fullscreen state directly in ShellClient
So far we delegated to ShellSurfaceInterface, but it's possible that
we call into isFullscreen when the ShellSurfaceInterface is already
destroyed.

Note: the functionality needs to be slightly reworked. The fullscreen
state should change once we get a buffer for fullscreen, not when the
client requests it. This is a general pattern which is wrong also for
maximized, etc.
2015-10-08 10:05:40 +02:00
Martin Gräßlin
d25c465211 Move layer functionality to AbstractClient
Merges together the code from ShellClient and Client and removes the
starting differences. Long term it's better to have only one
implementation to prevent diversions in the implementation.

As it doesn't match exactly protected virtual methods are called
which allow more specific implementations for a certain aspect of the
layer resolving.
2015-09-17 11:06:59 +02:00
Martin Gräßlin
a86a7e7b3c Add a placement strategy for transient ShellClients
A transient ShellClient has an offset position to the parent surface.
Use this to position the ShellClient properly.

This fixes the random placement of menus.
2015-09-16 13:52:25 +02:00
Martin Gräßlin
0935a60b38 Implement transientFor in ShellClient 2015-09-16 13:52:25 +02:00
Martin Gräßlin
79317717c8 [wayland] Export ShellClient
Need to access it from our QPA plugin.
2015-08-25 14:33:50 +02:00
Martin Gräßlin
c56bbc0ddf Composite windows from a QOpenGLFramebufferObject
This change introduces a mechanism for internal windows to be rendered
to a QOpenGLFramebufferObject to be composited using the texture bound
to the FBO. This is useful for in-process rendering (e.g. QtQuick) and
at the same time bypassing the windowing system.

The OpenGL context of the QOpenGLFramebufferObject needs to be sharing
with the compositing OpenGL context.
2015-08-25 14:33:50 +02:00
Martin Gräßlin
a5185f924d Fix naming 2015-07-22 08:38:00 +02:00
Martin Gräßlin
8d407157c4 Provide a base implementation for performMouseCommand in AbstractClient
Most commands can be operated directly in AbstractClient already which
means less code duplication to make the commands work for wayland
clients.
2015-07-09 16:34:19 +02:00
Martin Gräßlin
72635101f0 [wayland] Improve creation of KWayland::Server::PlasmaWindowInterface
The creation of PlasmaWindowInterface is moved from WaylandServer into
AbstractClient. This allows the sub classes to better control when to
create/destroy the Client.

For creation it's bound to becoming visible - that is Windows which are
only created but never shown are not announced at all.

For Client it's destroyed with the normal tear-down of a Client, for
ShellClient it's destroyed on unmapped (which also means a new one
will be created again in case of another mapping of the surface).

As a side effect, this works around the problem that ShellClients do not
yet get destroyed for QtWayland's menus (needs further investigation).
2015-07-09 09:10:33 +02:00
Martin Gräßlin
da4305c2fa [wayland] Focus to null X11 window when activating a Wayland client
Improves interaction with Xwayland clients.
2015-07-07 16:15:58 +02:00
Martin Gräßlin
31e599a63d [wayland] Provide icon from ShellSurface's windowClass
We can use the window class to locate a desktop file and from this
we can load a themed icon.
2015-06-20 01:11:42 +02:00
Martin Gräßlin
2192bd69e0 [wayland] Add support for PanelBehavior in ShellClient
The PanelBehavior determines the stacking of panel surfaces and whether
they "have a strut".
2015-06-20 00:14:49 +02:00
Thomas Lübking
39c35f8f48 fix sendToScreen/checkWorkspaceGeometry
Prime target is to preserve the in-screen
condition of client AND window.i[1]
Atm. when the client is fully in sight
(but the window is not) - regardless of snapping
or screen change - a workspace update (screen change,
resolution change, adding/removing a strutting panel)
would allow the client to partiall escape screen bounds.

This is changed so that if the client is fully in sight,
it's kept fully in sight (but not the decoration)
If the entire window was fully in sight, it's also kept
(as is right now)
The code handles inner screen edges (if the client was in sight,
the entire window will be if we'd bleed to the other screen)

[1] I'd say that handling the client is more relevant,
but foresee more complaints if the window wasn't handled anymore ;-)

During that, i stumbled across some other issues.
- when a window centered on one screen is moved to a screen smaller
  than the window, the window is shrinked to the dimensions of that
  screen and now randomly touches eg. left AND right edge. When
  moved back, the right and bottom edge were preferred
  (to the window was "moved" into the lower right corner).
  It's now kept centered.
- geom_restore was saved before keeping the window in the new
  screen area (causing accidental drops on screen changes)

BUG: 330968
REVIEW: 122517
FIXED-IN: 5.4

Prime target is to preserve the in-screen condition
of client AND window.[1]
Atm. when the client is fully in sight (but the window is not) -
regardless of snapping or screen change - a workspace update
(screen change, resolution change, adding/removing a strutting
panel) would allow the client to partiall escape screen bounds.

This is changed so that if the client is fully in sight,
it's kept fully in sight (but not the decoration)
If the entire window was fully in sight, it's also kept
(as is right now)
The code handles inner screen edges (if the client was in sight,
the entire window will be if we'd bleed to the other screen)

[1] I'd say that handling the client is more relevant,
but foresee more complaints if the window wasn't handled anymore ;-)

During that, i stumbled across some other issues.
- when a window centered on one screen is moved to a screen
smaller than the window, the window is shrinked to the dimensions
of that screen and now randomly touches eg. left AND right edge.
When moved back, the right and bottom edge were preferred
(to the window was "moved" into the lower right corner).
It's now kept centered.
- geom_restore was saved before keeping the window in the new
screen area (causing accidental drops on screen changes)

BUG: 330968
REVIEW: 116029
FIXED-IN: 5.3
2015-06-19 19:48:39 +02:00
Thomas Lübking
41be18e317 Robust modal activation; ensure showing activated
...window

REVIEW: 123875
BUG: 348083
2015-06-10 08:50:16 +02:00
Martin Gräßlin
f696b578cc [wayland] Add support for QtSurfaceExtension for closing ShellClients
If the ShellClient has a QtExtendedSurface it's closeable.
2015-06-10 00:59:53 +02:00
Martin Gräßlin
124bd8aaed [wayland] Add support for the PlasmaShell interface
The PlasmaShell interface allows to create a PlasmaShellSurface for a
given Surface. Through this interface the Surface can request:
* a specific position
* a window type

So far only the window types Normal, Panel and Desktop are supported
which is a sufficient subset for getting plasmashell to work.

In future there should be security checks so that only the dedicated
desktop shell can bind these interfaces.
2015-06-09 19:10:56 +02:00
Martin Gräßlin
280075b7cd [wayland] Support windowType for internal windows
Our internal windows (e.g. a declarative script) can specify their own
window type which then can get honored by KWin.
2015-06-08 21:27:55 +02:00
Martin Gräßlin
7cded49145 [wayland] Implement maximizing of ShellClient 2015-06-07 03:28:38 +02:00
Martin Gräßlin
7f0758b85f [wayland] Handle pointer events for internal windows
If there is a visible internal window it gets the pointer events.
The assumption is that the last created internal window is the top
most in stacking order.
2015-06-06 17:43:13 +02:00
Martin Gräßlin
35abacabc4 [wayland] Handle unmap in ShellClient
Do not destroy the ShellClient when the Surface gets unmapped. Instead
just hide the ShellClient and show it again on the next damage.
2015-06-03 21:19:00 +02:00
David Edmundson
9791978d53 enable wayland clients to go fullscreen
REVIEW: 123853
2015-05-28 10:58:40 +01:00
Martin Gräßlin
007e1253c6 Add resizeWithChecks to AbstractClient
The virtual method does not take the xcb_gravity_t - that's only added
in a variant in Client.
2015-05-27 14:20:11 +02:00
Martin Gräßlin
da2731be51 Move Client::move to AbstractClient and add implementation in ShellClient 2015-05-27 10:16:46 +02:00
David Edmundson
df1499784f move isFullScreenable to AbstractClient
REVIEW: 123871
2015-05-21 16:40:21 +01:00
Martin Gräßlin
45c1eddf4a Make Toplevel::window() virtual and implement in ShellClient
We need a shared window id in some way to e.g. properly support TabBox.
Though I do not like that the return value is xcb_window_t. This should
be either changed or a new "generic" windowId needs to be introduced.
2015-05-21 13:34:26 +02:00
Martin Gräßlin
23b19b4efe [wayland] Add a bool ShellClient::isInternal
Helper method to identify whether a ShellClient is for one of KWin's
internal connections (e.g. a QWindow).
2015-05-21 13:34:26 +02:00
Martin Gräßlin
2ff903e8b3 [wayland] Sync ShellClient's geometry with internal QWindow's geometry
Update geometry when the referenced internal QWindow changes position.
This makes the close button in PresentWindows work.
2015-05-20 08:27:48 +02:00
Martin Gräßlin
7d152991c7 [wayland] ShellClient can reference an internal QWindow
If the ShellClient got created for a Qt internal window, we try to
find the QWindow and if we get one, we use the geometry directly as
it got set by KWin in the first place.

Also a windowId() is added to ShellClient which can be used by the
effect system to find an EffectWindow. If it's an internal QWindow
we just use that window id. For other clients we still need some
smart solution.
2015-05-18 15:31:41 +02:00
Martin Gräßlin
e9e055dcca Call Workspace::clientHidden when ShellClient gets destroyed
Ensures that active_client gets changed if it was a ShellClient.
2015-05-08 12:43:47 +02:00
Martin Gräßlin
912ab71abd ShellClient inherits AbstractClient
This allows to make ShellClient behave more like a Client and allows
to activate it. Some pieces for activating are still missing.
2015-05-08 12:43:47 +02:00
Martin Gräßlin
b6f793090e Toplevel::setOpacity becomes virtual
It calls unconditionally into NETWinInfo.
2015-05-08 12:43:47 +02:00
Martin Gräßlin
d60c377890 [wayland] Introduce a ShellClient
The ShellClient is a Toplevel subclass for a
KWayland::Server::ShellSurfaceInterface. It gets created when a new
ShellSurfaceInterface is created and destoryed when it gets unmapped.

So far the usage is still rather limited. The ShellClient is opened
at position (0/0). While it's possible to pass pointer events to it,
it's not yet possible to activate it, so no keyboard focus.
2015-05-08 12:43:44 +02:00