Add a new object class, MetaWaylandInputDevice, and make all device
classes (pointer, keyboard, touch) inherit it. In the future common
functionality may be placed there.
https://bugzilla.gnome.org/show_bug.cgi?id=771305
Meant to replace explicitly checking whether a
MetaWaylandPointer/MetaWaylandKeyboard/MetaWaylandTouch has a seat or
not to determine whether they are supposed to be active or not.
https://bugzilla.gnome.org/show_bug.cgi?id=771305
Since Xwayland surface constraints might need to enable not only
because the constrained window appears focused, add a pointer focus
listener and try constrain whenever the pointer focus changes. It's
still required that a Xwayland window is focused to activate.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Xwayland surfaces are special, because there is no reliable way to
associate a window with its corresponding "application window" (the one
which was given focus). Many games that require pointer warping and
confining pointer grabs may for example create override redirect windows
and make that window receive input even though it will never be the
focus window.
Therefore, the requirements for enabling a constraint for a wl_surface
from Xwayland needs to be relaxed in order. This commit changes
Xwayland wl_surfaces to not require being focused to be enabled; it'll
be enabled as long as any X11 window is the one with focus.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Require pointer focus to enable, otherwise we can't guarantee it has
entered the surface, as the focus may have been given to a subsurface,
override-redirect or other sub window covering the surface that was
requested to have o pointer constraint.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Add a signal that is emitted when the pointer focus surface of the
pointer device changes. This will later be used by the pointer
constraints to maybe enable pointer constraints when a surface receives
pointer focus.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
For Xwayland, a newly created wl_surface and X11 Window pair may not be
immediately associated, but Xwayland may still request a pointer
constraint on some of its wl_surface's. Handle the situation by
postponing maybe enabling the constraint until the window and surface
has been associated.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
Move the MetaWaylandSurface::destroy signal before starting the actual
destruction, in wl_surface_destructor, so that all fields (e.g. surface
role) are intact when the listeners are invoked.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
When the Xwayland wl_surface is created, it may not yet be possible to
associate it with the corresponding X11 Window. Add a signal to the
Xwayland role to communicate with any interested parties.
https://bugzilla.gnome.org/show_bug.cgi?id=771050
If a client would attach a buffer to a surface, commit, destroy the
buffer and then later set the surface as a cursor, there will be no
wl_buffer available to be used by the cursor role. Instead of
dereferencing the non-existing wl_buffer resource, handle this situation
by logging a warning and treating a prematurely destroyd wl_buffer as if
no buffer had been attached.
https://bugzilla.gnome.org/show_bug.cgi?id=770992
Implement min/max size request from xdg-shell-v6 and plug it into the
existing code so that windows with fixed size cannot be tiled/maximized
in Wayland just like in X11.
Bugzilla: https://bugzilla.gnome.org/show_bug.cgi?id=770226
The seat capability updating is synchronous, but input events are
asynchronous (first queued then emitted). This means we may end up in a
situation where we from libinput first may receive a key event,
immediately followed by a device-removed event. Clutter will first
queue the key event, then remove the device, immediately triggering the
seat capability removal.
Later, when the clutter stage processes the queued events, the
previously queued key event will be processed, eventually making it
into MetaWaylandSeat. Before this patch, MetaWaylandSeat would still
forward the key event to MetaWaylandKeyboard, even though it had
'released' it. Doing this would cause referencing potentially freed
memory, such as the xkb state that was unreferenced when the seat
removed the capability.
In order to avoid processing these lingering events, for now, just drop
them on the floor if the capability has been removed.
Eventually, the event queuing etc needs to be redesigned to work better
when used in a Wayland compositor, but for now at least don't access
freed memory.
https://bugzilla.gnome.org/show_bug.cgi?id=770727
We may be assigned multiple times, if the surface is assigned to be a
cursor surface multiple times. Each time e.g. wl_pointer.set_cursor is
called, we'll be assigned.
While the role object exists, we'll handle buffer use count even when
we are not actively assigned, thus we should only handle the initial
assignment use count bump when constructing, so that we don't increase
it when reassigned, where the wl_resource may already have been
released.
https://bugzilla.gnome.org/show_bug.cgi?id=770402
For backend handled cursors, if nothing else changes on the clutter
stage, we end up not sending out frame callbacks since clutter doesn't
draw a new frame.
To fix this, we'll keep cursor surfaces' frame callbacks separate from
other surfaces' and trigger them from the new
MetaCursorRenderer::cursor-painted signal which handles both software
and hardware cursors.
https://bugzilla.gnome.org/show_bug.cgi?id=749913
If cogl fails to create a texture from the client's given buffer,
mutter would raise a fatal error and terminate.
As a result, a broken client might kill gnome-shell/mutter and take the
entire Wayland session with it.
Instead of raising a fatal error in this case, log the cogl error
message and send the client an OOM error, so mutter/gnome-shell can
survive an unsupported buffer size.
https://bugzilla.gnome.org/show_bug.cgi?id=770387
Windows from Xwayland still needs to use the Wayland path, but is
represented an MetaWindowX11, thus the abstraction introduced in
"window: Make meta_window_has_pointer() per protocol implemented"
is wrong. Lets turn back time, and reconsider how this can be
abstracted more correctly in the future.
This reverts commit 9fb891d216.
Rely on the actor surface role's commit function for queuing frame
callbacks. This also makes the surface actor state synchronization work
again, which was broken by 'wayland: Sync surface actor state in actor
role commit handler'.
https://bugzilla.gnome.org/show_bug.cgi?id=770131
Don't check whether the surface of the role has a window, but whether
the corresponding toplevel surface has a window. This is necessary to
make subsurfaces not always early out.
https://bugzilla.gnome.org/show_bug.cgi?id=770131
There is nothing stopping a subsurface from commiting its state before
its parents role has been assigned. Thus, we need to handle
meta_wayland_surface_get_toplevel() returning NULL for subsurfaces even
on commit.
Make sure to always call the parent role commit vfunc, so that they can
handle updating their state properly.
This means other places need to handle the situation where
surface->window is NULL on commit. This may for example happen when the
parent of a modal dialog is unmapped or NULL is attached to a
wl_shell_surface.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Port the xdg_shell implementation to use the unstable v6 protocol. This
includes:
- making xdg_surface a generic base interface for xdg_shell surface
roles
- create a xdg_toplevel role replacing the old xdg_surface
- change the xdg_opup role to be based on xdg_surface
- make xdg_popup not grab by default
- add support for xdg_positioner
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Emit a 'configure' signal before configuring the role. This will enable
extensions to send its own configure events before the role is
configured.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Add support for assigning a window a custom window placement rule used
for calculating the initial window position as well as defining how a
window is constrained.
The custom rule is a declarative rule which defines a set of parameters
which the placing algorithm and constrain algorithm uses for
calculating the position of a window. It is meant to be used to
implement positioning of menus and other popup windows created via
Wayland.
A custom placement rule replaces any other placement or constraint
rule.
https://bugzilla.gnome.org/show_bug.cgi?id=769936
Allow passing parameters (only GObject parameters supported for now) so
that role assignment can affect the paremeters set during construction.
If a role was already assigned when assigning, the passed parameters
are set using g_object_set_valist().
https://bugzilla.gnome.org/show_bug.cgi?id=769936
This commits adds support for exporting xdg_surface handles via
xdg_exporter and importing them via xdg_importer.
This bumps the required wayland-protocols version to 1.6.
https://bugzilla.gnome.org/show_bug.cgi?id=769786
Meant to be used by users of MetaWaylandSurface's that need to know
when the surface was unmapped. So far only emitted by shell surfaces
(surfaces with MetaWindow's).
https://bugzilla.gnome.org/show_bug.cgi?id=769786
We do some things when binding to a socket fails (closing the fd,
logging, unlinking files, ...) those might affect errno in some
or other way, so it might no longer be EADDRINUSE even if we later
try to make those non fatal.
It seems better to check errno soon after the failure, and don't
rely on it in any way at a later point. All error paths in
bind_to_abstract_socket() also have early logging, which also might
help figure out better the point of failure when the socket fails
to be created.
https://bugzilla.gnome.org/show_bug.cgi?id=769578
We only use a handful of the attributes set, so lets stop pretending
that things are initialized for a reason. Eventually we should stop
using XWindowAttributes in the generic MetaWindow creation path.
https://bugzilla.gnome.org/show_bug.cgi?id=769070
If the compiler cannot figure out that the condition for setting
the dev variable is the same as the condition for accessing it,
it will complain about potential uninitialized use.
We must lookup the mode switch serial for the group where the button
belongs to. Also, avoid the changes if the client requests setting
the feedback for buttons owned by the compositor.
We assumed that each group could only have 1 strip and/or ring, because
accounting is performed per group, so we could not assume the real
index for anything above 1. Get rid of this restriction, now that
MetaWaylandTabletPad does its own accounting of rings/strips, alongside
groups.
This is best for 2 reasons:
- It's feels cleaner doing first creation of rings/strips and then
the group assignment. The other option is making groups iterate
other all rings/strips and selectively skip those not meant for
it, which sounds somewhat redundant.
- Some minimal accounting of rings/strips without group restrictions
is needed for meta_wayland_tablet_pad_get_label().
The rings/strips memory is now owned by MetaWaylandTabletPad instead
of groups, which is sort of meaningless since all are meant to go
at the same time.
All pads will share the same focus than the keyboard, so this means that:
- The focus changes in-sync for keyboard and all pad devices, and
- Newly plugged pads will be immediately focused on that same surface
This object represents the collection of buttons, strips and rings
in a tablet pad. All the objects created (pad, strips and rings)
share a common focus surface and have the same lifetime.
This is now separated from the generic cursor one. This means that wl_surfaces
can't be shared across wl_pointer and wp_tablet_tool. This is a change in
tablet protocol v2.
This is a simple subclass of MetaWaylandSurfaceRoleCursor, mostly
so we can distinguish by GType, the methods in the parent class
still apply and are useful.
Using clutter API to transform coordinates is only accurate right
after a clutter layout pass but this function is used e.g. to deliver
pointer motion events which can happen at any time. This isn't a
problem for wayland clients since they don't control their position,
but X clients do and we'd be sending outdated coordinates if a client
is moving a window in response to motion events.
https://bugzilla.gnome.org/show_bug.cgi?id=768039
The call fetching the targets mistakenly used the timestamp meant
to back up the TIMESTAMP atom (hence, it's the timestamp at which
the selection is *owned* by the compositor, on behalf of a wayland
client).
This timestamp is actually only updated when the compositor gets
to own the selection, so it's a randomly late timestamp to retrieve
the TARGETS atom content, which certain clients might end up
ignoring.
https://bugzilla.gnome.org/show_bug.cgi?id=768007
The result flag needs to be marked as moved even for pending moves,
otherwise the window's unconstrained_rect doesn't get updated in
meta_window_move_resize_internal() and the anchor grab is wrong.
https://bugzilla.gnome.org/show_bug.cgi?id=764180
The X11 backend uses EWMH's _NET_WM_PID to get the PID of an offending
client and kill its PID to force the client to terminate.
The Wayland backend is using a Wayland protocol error, but if the client
is hung, that will not be sufficient to kill the client.
Retrieve the client PID under Wayland using the Wayland client API
wl_client_get_credentials() and kill() the client the same way the X11
backend does.
https://bugzilla.gnome.org/show_bug.cgi?id=767464
xdg-shell allows desktop environments to extend the list of states
within a given range.
Use this possibility to add a new state for tiled so that gtk+ can
benefit from this.
https://bugzilla.gnome.org/show_bug.cgi?id=766860
This makes us behave the same both on bind and when an output
changes. In particular, we were not sending scale and done events on
output changes. We were also unconditionally sending mode events on
output changes even though these should only be sent if there is an
actual mode change.
https://bugzilla.gnome.org/show_bug.cgi?id=766528
If we try to send notify event (either from surface_state_changed()
or from meta_window_wayland_move_resize_internal()),
we will crash, because we don't have a sufrace anymore.
There's no reason why to resize the window that is being
unmanaged anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=751847
meta_wayland_tablet_manager_update()/handle_event() are called before
the MetaWaylandSeat counterparts. If the event comes from a device
managed by MetaWaylandTabletManager, the event will be exclusively handled
by it.
Each tool has its own MetaCursorRenderer instance, which is created/destroyed
upon proximity, and possibly updated through focus and set_cursor calls in
between.
This struct keeps the server side information for the wl_tablet_manager
global resource. It keeps the clients requesting this interface, and
does keep track of the plugged tablet devices, so
wl_tablet_manager.device_added is emitted on the expected clients.
Move into a standalone meta-wayland-surface-role-cursor.[ch], and
make generic enough to work for pointe and additional (eg. tablet)
cursors.
Most notably, the sprite is now kept completely internal to the
cursor role, and updates are routed through the given
MetaCursorRenderer (which may be the default one for the pointer,
or something else).
The way cursor updates after cursor surface destruction has also
been reworked, the pointer will just keep track of the last cursor
surface, so older surfaces being destroyed don't trigger pointer
rechecks/updates.
There's places where it would be convenient to add listeners to this,
so add the signal. The signal is only emitted once during destruction,
it is convenient for the places where we want notifications at a time
the object is still alive, as opposed to weak refs which notify after
the fact.
Separate "xdg_surface", "xdg_popup" and "xdg_shell" related functions
into three sections. Prior to this, the "xdg_shell" part was a bit all
over the place.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Dismiss the popup when the parent is destroyed, and do this in the
destructor of the parent object. This makes the parent destory listener
unnecessary, since we already handle the parent child unlinking
explicitly in the object destructor.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Instead of relying on destroy signals attached to the corresponding
role object, let the roles explicitly dismiss the popup when it should
be dismissed.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Add a bridge between the MetaWaylandPopup object and the corresponding
popup surface role. This bridge replaces communicating dismissed and
unmapped popup events.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Before this commit, on Wayland, the buffer rect would have the size of
the attached Wayland buffer, no matter the scale. The scale would then
be applied ad-hoc by callers when a sane rectangle was needed. This
commit changes buffer_rect to rather represent the surface rect (i.e.
what is drawn on the stage, including client side shadow). The users of
buffer_rect will no longer need to scale the buffer_rect themself to
get a usable rectangle.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
A large part of meta_wayland_surface_apply_window_state() was only
relevant for xdg_surface. Make this more obvious by splitting it up,
moving the relevant parts to the relevant roles.
https://bugzilla.gnome.org/show_bug.cgi?id=763431
Move xdg_shell related functionality to a new meta-wayland-xdg-shell.c
and wl_shell related functionality to a new meta-wayland-wl-shell.c,
and adapt role object tree.
Common functionality related to the surface being drawn as a
MetaSurfaceActor was moved to a MetaWaylandSurfaceRoleActorSurface role.
The subsurface role GObject is made to inherit the actor surface GObject.
Shell surface hooks (configure, ping, close, popup done) were added to
a MetaWaylandSurfaceRoleShellSurface GObject which inherits the
surface actor role GObject.
The shell surface roles (xdg_surface, xdg_popup, wl_shell_surface) are
made to inherit the shell surface GObject and implement the relevant
API.
https://bugzilla.gnome.org/show_bug.cgi?id=757623https://bugzilla.gnome.org/show_bug.cgi?id=763431
Leave these checks up to the callers, the only uses of this function
(indirect, through meta_wayland_seat_get_grab_info) are
[wl_shell|xdg]_surface.move/resize/show_window_menu.
In move/resize it makes sense to check for a button being pressed, because
we must expect a button release event. However for xdg_surface.show_window_menu
we 1) don't strictly need further events and 2) we must account for press+release
event pairs being processed at once in the compositor before the client sees
the former.
That is eg. the case of touchpad 2nd/3rd button tap emulation, multifinger
taps will emit the event pair at once, so when the client manages to request
xdg_surface.show_window_menu, it'll be too late in the compositor side, so the
request is ignored.
https://bugzilla.gnome.org/show_bug.cgi?id=764519
A wl_data_device object may be created while it is being focused,
either because the client destroyed it or because the client was
destroyed. Handle this by early out in focus handler vfuncs the case
where it was destroyed, so that we don't corrupt memory and/or cause
segmentation fault.
https://bugzilla.gnome.org/show_bug.cgi?id=765062
If we get a key event but still have pending modifier state changes we
need to send a modifiers event right away so that the key event can be
interpreted by clients correctly modified.
This case could happen when mutter/gnome-shell itself consumes the
modifier key press event such as with the overview key which by
default is triggered on super press.
https://bugzilla.gnome.org/show_bug.cgi?id=748526
The wayland protocol has enough space to send both virtual and real
modifiers on modifiers events which saves clients the work of
resolving virtual modifiers themselves.
https://bugzilla.gnome.org/show_bug.cgi?id=748526
While CoglError is a define to GError, it doesn't follow the convention
of ignoring errors when NULL is passed, but rather treats the error as
fatal :-(
That's clearly unwanted for a compositor, so make sure to always pass
an error parameter where a runtime error is possible (i.e. any CoglError
that is not a malformed blend string).
https://bugzilla.gnome.org/show_bug.cgi?id=765058
Just like we do for _NET_WM_MOVERESIZE messages on X11, consider
wayland client move/resizes as "frame actions" so that the same
constraints are applied to them, in particular the titlebar visibility
constraint.
https://bugzilla.gnome.org/show_bug.cgi?id=748819
In order for the native cursor renderer to be able to create a hw
cursor in response to wl_pointer.set_cursor(), keep a private use-count
and reference to the active buffer, stopping it from being released
until it is consumed, replaced, or the surface is destroyed.
https://bugzilla.gnome.org/show_bug.cgi?id=762828
Whether a surface needs to keep the committed wl_buffer un-released
depends on what role the surface gets assigned to. For example a cursor
role may need an unreleased shm buffer in order to create a hw cursor from
it.
In order to support this, keep a separate reference and use count to
the buffer on behalf of the in the future assigned role, and release
those references after the surface was assigned a role. A role that
needs its own references and use counts, must in its assign function
make sure to add those.
https://bugzilla.gnome.org/show_bug.cgi?id=762828
Each wl_surface.commit with a newly attached buffer should result in
one wl_buffer.release for the attached buffer. For example attaching
the same buffer to two different surfaces must always result in two
wl_buffer.release events being emitted by the server. The client is
responsible for counting the wl_buffer.release events and be sure to
have received as many release events as it has attached and committed
the buffer, before reusing it.
https://bugzilla.gnome.org/show_bug.cgi?id=762828
This target is set whenever DnD moves towards an area between surfaces.
Although no offer is set and data is actually not read, drag sources
offering this mimetype will be able to behave just like they used to
do in X11.
https://bugzilla.gnome.org/show_bug.cgi?id=762104
We want some initial processing, even if the current focus didn't change.
This could be for example the case of starting DnD too close to the window
edge and out of it. At the point start_drag() is called, the current
pointer focus is already NULL, so set_focus() would simply bail out here.
https://bugzilla.gnome.org/show_bug.cgi?id=762104
On the X11 backend we don't track the pointer position in
priv->current_x/y which remain set to zero. That means we never set
the clutter stage cursor if point 0,0 isn't covered by any monitor
since we return early.
Commit 4bebc5e5fa introduced this to
avoid crashing on the prepare-at handlers when the cursor position
doesn't fall inside any monitor area but we can handle that higher up
in the stack. In that case, the sprite's scale doesn't matter since
the cursor won't be shown anyway so we can skip setting it.
https://bugzilla.gnome.org/show_bug.cgi?id=763159
This commits adds a gtk_surface.present request and its implementation.
The timestamp is assumed to be from some input event that the client
responded to. The timestamps we deal with when managing windows will
usually come from two different clocks: CLOCK_MONOTONIC if they come
from libinput/evdev, or CLOCK_MONOTONIC_COARSE if they come from the
X server.
Luckily these are quite similar, the difference beeing that the X server
timestamps having lower resolution, so we can just pass the timestamps
no matter where they came from and it'll most likely work fine, except
for the race condition described in bug 756272 which might happen here
too until it is properly fixed.
https://bugzilla.gnome.org/show_bug.cgi?id=763295
Add a system_bell request to gtk_shell. A client can use this to invoke
the system bell, be it aural, visual or none at all. Currently per
window visual bell support is not implemented.
https://bugzilla.gnome.org/show_bug.cgi?id=763284
The gtk_shell protocol used some half baked unstable protocol semantics
that worked by only allowing binding the exact version of the
interface. This hack is a bit too confusing and it makes it impossible
to do any compatible changes without breaking things.
So, instead rename it to include a number in the interface names. This
way we can add requests and events without causing compatibility issues,
and we can later remove requests and events by bumping the number in
the interface names.
https://bugzilla.gnome.org/show_bug.cgi?id=763284
Using the current position to set the origin x/y of the DnD icon
is wrong, it should still be used in order to move the icon besides
the current pointer position though.
Fixes possible drag-start-x/y property constraint warnings when
starting a drag too close to the window edge, and towards outside
of it.
If a MetaWindow's 'appears-focused' state changed to true, but the
window did not have pointer focus, the constraint did not enable. Thus,
make it possible for the user to also click the window to enable it.
https://bugzilla.gnome.org/show_bug.cgi?id=762661
Instead of relying on the keyboard focus surface, use the
'appears-focused' state of the corresponding MetaWindow to determine if
a constraint should enable or not.
https://bugzilla.gnome.org/show_bug.cgi?id=762661
Instead of having MetaWindowWayland having hooks into pointer
constraints subsystem, have the pointer constraints subsystem listen
for the signal itself and enable/disable itself.
https://bugzilla.gnome.org/show_bug.cgi?id=762661
Instead of having a very large region represent an infinitely large
region, use NULL, and use the calculated input region from the
MetaWaylandSurface if the constraint region was not set.
https://bugzilla.gnome.org/show_bug.cgi?id=762661
The when surface->input_region is NULL, it should be interpreted as the
whole surface region. If not, the effective input region is the
intersection of the buffer region and the input region set by
wl_surface.set_input_region. Add
meta_wayland_surface_calculate_input_region() that does this
calculation.
https://bugzilla.gnome.org/show_bug.cgi?id=762661
If we receive multiple SelectionRequest events, we'll end up replacing the
former WaylandSelectionData at a time when an async read has been issued.
This will cause the cancellation of the previous operation.
But the wayland_data_read() callback will attempt to just remove the
current wayland data again on error, which will not be the one we're
cancelling, so the new operation will just be cancelled too.
Also, cancellation is no longer warned about. As the wayland selection
has been replaced at this time, we can just return here.
https://bugzilla.gnome.org/show_bug.cgi?id=760745
This is necessary for the X11 side to catch up, and unset the primary
selection ownership on our window that represents the wayland side in
X11 selection.
https://bugzilla.gnome.org/show_bug.cgi?id=760745
This is necessary for the X11 side to catch up, and unset the selection
ownership on our window that represents the wayland side in X11 selection.
https://bugzilla.gnome.org/show_bug.cgi?id=760745
We may have released the wl_buffer already when doing this, which means
we should not try to access the wl_buffer content.
Regarding the cursor texture this is not an issue since we can just use
the texture created in apply_pending_state().
The hw cursor however will only be realized if the surface is already
using the the buffer (surface->using_buffer == true). This will, at the
moment, effectively disable hardware cursors for SHM buffers.
https://bugzilla.gnome.org/show_bug.cgi?id=762828
Don't unset the surface->buffer if the associated wl_buffer object is
destroyed. The MetaWaylandBuffer doesn't really only represent a
wl_buffer object, but also the data (texture) created from the given
wl_buffer. Thus, for example destroying a released SHM wl_buffer should
not destroy the MetaWaylandBuffer instance, because the texture may
still be used.
This commit also fixes a race where calc_showing would hide a window
because, at the time of calculation whether it should be showing, the
surface's buffer had been destroyed as described above.
https://bugzilla.gnome.org/show_bug.cgi?id=762716
In destroy_data_offer() there is code to ensure compatibility when
dragging from a v3 wl_data_device to a v2 one, it's however not checking
correctly that this is the DnD drag source. The other path should be
used otherwise.
https://bugzilla.gnome.org/show_bug.cgi?id=762878
Add an additional MetaWaylandDataSource implementation for primary selection
sources, and methods to set primary selection offers. Primary selection
sets altogether a different channel than the clipboard selection, those don't
cross in any way.
Also, the bridge for the X11 PRIMARY selection atom has been added, which
adds all the necessary handling to translate primary selection both ways
with wayland and X11 applications.
https://bugzilla.gnome.org/show_bug.cgi?id=762560
This protocol is an internal mirror of the primary selection drafts
being proposed for wayland-protocols. No changes besides prefix/suffix
changes.
https://bugzilla.gnome.org/show_bug.cgi?id=762560
The constraint may be destroyed before the client destroyes the
protocol object, for example if a oneshot constraint was disabled by
alt-tab. Therefore we need to NULL check the constraint in request
handlers and ignore any requests to defunct objects.
Add a gtk_shell.set_startup_id request, so the application can communicate
to the compositor the startup id that it received through the
DESKTOP_STARTUP_ID envvar, or other means.
https://bugzilla.gnome.org/show_bug.cgi?id=762268
Since a buffer can be used by multiple surfaces at once,
we need to release the buffer only after all surfaces
are finished with it. Currently we track whether or
not to release the buffer based on the accessible boolean.
This commit changes it to a counter to accomodate multiple
users.
Also, each surface needs to know whether not it is done with
the buffer, so this commit adds a buffer_used boolean to the
surface state.
https://bugzilla.gnome.org/show_bug.cgi?id=761613
We currently track whether or not a buffer can be released early
by looking at the copied_data boolean on the buffer. This boolean
is, practically speaking, always set to TRUE for shm buffers and is
always false otherwise.
We can just as easily check if the buffer is a shm buffer to decide
whether or not to do an early release. That's better from a
theoretical point of view since copied_data assumes a 1-to-1
relationship between surface and buffer, which may not actually hold.
This commit drops copied_data and changes the check to instead see
if the buffer is shm.
https://bugzilla.gnome.org/show_bug.cgi?id=761613
meta_wayland_surface_toplevel_commit has a lot of logic to handle
a new buffer getting attached as part of the commit. None of
that code needs to run if there is no new buffer attached.
This commit short-circuits that case.
https://bugzilla.gnome.org/show_bug.cgi?id=761613
This patch adds support for confinement regions that are more complex
than a single rectangle. It relies on details about cairo regions not
explicitly in the API in order to generate the outer border of the
region.
https://bugzilla.gnome.org/show_bug.cgi?id=744104
The x/y coordinates of the ClutterInputDevice were not the ones which was
the result of this event but whatever event was queued the last. The
correct coordinates can, however, be found in the event itself, so lets
use those.
https://bugzilla.gnome.org/show_bug.cgi?id=744104
The wp_pointer_constraints protocol is a protocol which enables clients
to manipulate the behavior of the pointer cursor associated with a seat.
Currently available constraints are locking the pointer to a static
position, and confining the pointer to a given region.
Currently locking is fully implemented, and confining is implemented for
rectangular confinement regions.
What else is lacking is less troublesome semantics for enabling the lock
or confinement; currently the only requirement implemented is that the
window that appears focused is the one that may aquire the lock.
This means that a pointer could be 'stolen' by creating a new window that
receives active focus, or when using focus-follows-mouse, a pointer
passes a window that has requested a lock. This semantics can be changed
and the protocol itself allows any semantics as seems fit.
https://bugzilla.gnome.org/show_bug.cgi?id=744104
Add support for sending relative pointer motion deltas to clients who
request such events by creating wp_relative_pointer objects via
wp_relative_pointer_manager.
This currently implements the unstable version 1 from wayland-protocols.
https://bugzilla.gnome.org/show_bug.cgi?id=744104
The code currently implements a function, get_time, that
fetches a timestamp. That duplicates code already in glib,
and the glib implementation is better, anyway, since it doesn't
skew backward when the system clock is changed.
This commit changes the code to use g_get_monotonic_time and
drop the get_time function.
https://bugzilla.gnome.org/show_bug.cgi?id=761613
commit 0165cb6974 changed
mutter to release committed shm buffers as soon as they were
uploaded to the GPU.
It also inadvertently changed mutter to prematurely
release EGL buffers (which never get copied, but get used
directly).
This commit corrects that mistake.
https://bugzilla.gnome.org/show_bug.cgi?id=761312
When a client is ready for the compositor to read a surface's
shared memory buffer, it tells the compositor via
wl_surface_commit.
From that point forward, the baton is given to the compositor:
it knows it can read the buffer without worring about the client
making changes out from under it.
After the compositor has uploaded the pixel contents to the video
card it is supposed to release the buffer back to the client so that
the client can reuse it for future use.
At the moment, mutter only releases the buffer when a new buffer
is attached. This is problematic, since it means the client has
to have a second buffer prepared before the compositor gives the
first one back. Preparing the second buffer potentially involves
copying megabytes of pixel data, so that's suboptimal, and there's
no reason mutter couldn't release the buffer earlier.
This commit changes mutter to release a surface's buffer as soon
as it's done processing the commit request.
https://bugzilla.gnome.org/show_bug.cgi?id=761312
Since we are using the surface actor to draw the DND icon, the offset
is already accounted for by MetaSurfaceActorWayland, and passing the
surface position offset would effectively double the actual offset,
causing the icon to be misplaced.
This patch always sets the anchor offset to (0, 0) when the icon is a
Wayland surface, and lets the surface actor deal with the offsetting.
https://bugzilla.gnome.org/show_bug.cgi?id=759222
We now additionally send:
- wl_data_offer.source_actions
- wl_data_source.action
- wl_data_offer.action
- wl_data_source.dnd_drop_performed
- wl_data_source.dnd_finished
The protocol changes allow for compositors to implement different policies
when chosing the action, mutter uses this to reimplement the same behavior
that GTK+ traditionally had:
- Alt/Control/Shift modifiers change the chosen action to
ask/copy/move respectively
- Drags with middle button start out as "ask" by default
As mutter now also grabs the keyboard and unsets the window focus for these
purposes, the window focus is restored after the drag operation has
finished.
The Xdnd bridge code is also modified to cope with actions, so mixed
wayland-x11 scenarios are able to convey that information.
https://bugzilla.gnome.org/show_bug.cgi?id=760805
This will be useful during DnD, where mutter is expected to consume
keyboard events for either allowing changes in the selected DnD action,
or misc a11y features like keyboard-driven DnD.
Currently, the vtable contains 2 functions, key() will be used on every
key event we get from Clutter, modifiers() will notify of changes in the
keyboard modifiers (mouse buttons will never be set in the modifier mask)
https://bugzilla.gnome.org/show_bug.cgi?id=760805
As per the spec:
- wl_pointer.axis_source determines the current source of
scroll events.
- wl_pointer.axis_stop determines when there's no further
scroll events on the given axis.
- wl_pointer.axis_discrete is emitted on "wheel"
scroll sources, measured in ticks.
- wl_pointer.frame is meant to coalesce events that logically belong
together, e.g. axis events in this case.
Co-Authored-By: Peter Hutterer <peter.hutterer@who-t.net>
https://bugzilla.gnome.org/show_bug.cgi?id=760637
During xwayland initialization we run main loop and dispatch wayland
events, so that xwayland can initialize. If some client during this
phase connects and creates surface, mutter crashes because
it is not initialized yet. If we bind wayland socket after xwayland
is initialized and main loop is not running anymore, no client can
connect to mutter during initialization and that is what we want.
https://bugzilla.gnome.org/show_bug.cgi?id=751845
Use the xdg_shell XML file installed by wayland-protocols instead of
our own copy. This protocol has yet to go through any unstable naming,
but since we had an outdated (though wire compatible) version, some
minor changes were needed.
https://bugzilla.gnome.org/show_bug.cgi?id=758633
Remove our own copy of the pointer gestures protocol, and us the one
installed by wayland-protocols. This also means the new fixed unstable
naming conventions are used for the new version of the protocol, which
is reflected in the change. No functional changes were made, it is only
a rename.
https://bugzilla.gnome.org/show_bug.cgi?id=758633
When running as a native compositor, we can just do that. However, the
previous code must stay for whenever it's run as a X11 client.
Additionally, the fallback switch{} that transforms clutter 1-indexed
buttons into input.h event codes had to be adapted to the change introduced
in clutter commit 83b738c0e, where the 4-7 button range is kept clear for
compatibility with the X11 backend.
https://bugzilla.gnome.org/show_bug.cgi?id=758239
On the wire, Wayland specifies the refresh rate in milliHz. Mutter sends
the refresh rate in Hz, which confuses clients, e.g. weston-info:
interface: 'wl_output', version: 2, name: 4
mode:
width: 2560 px, height: 1440 px, refresh: 0 Hz,
flags: current preferred
interface: 'wl_output', version: 2, name: 5
mode:
width: 3200 px, height: 1800 px, refresh: 0 Hz,
flags: current preferred
and xrandr:
XWAYLAND0 connected 2560x1440+3200+0 600mm x 340mm
2560x1440@0.1Hz 0.05*+
XWAYLAND1 connected 3200x1800+0+0 290mm x 170mm
3200x1800@0.1Hz 0.03*+
Export the refresh rate in the correct units. For improved precision,
perform the KMS intermediate calculations in milliHz as well, and
account for interlaced/doublescan modes.
This is also consistent with what GTK+ expects:
timings->refresh_interval = 16667; /* default to 1/60th of a second */
/* We pick a random output out of the outputs that the window touches
* The rate here is in milli-hertz */
int refresh_rate = _gdk_wayland_screen_get_output_refresh_rate (wayland_display->screen,
impl->outputs->data);
if (refresh_rate != 0)
timings->refresh_interval = G_GINT64_CONSTANT(1000000000) / refresh_rate;
Where the 'refresh_rate' given is exactly what's come off the wire.
1000000000/60000 comes out as 16667, whereas divided by 60 is ...
substantially less.
https://bugzilla.gnome.org/show_bug.cgi?id=758653
Right now we just check the pointer serial, so the popup will be
immediately dismissed if the client passes a serial corresponding to
another input device.
Abstract this a bit further and add a meta_wayland_seat_can_popup() call
that will check the serial all input devices. This makes it possible to
trigger menus through touch or keyboard devices.
https://bugzilla.gnome.org/show_bug.cgi?id=756296
If we immediately dismiss the popup, we still need to set the
surface->xdg_popup pointer field in order for the destructor to
properly clean up the state. Not doing this may cause a crash if the
xdg_popup resource that was immediately dismissed is destoryed after
wl_surface during client destruction.
https://bugzilla.gnome.org/show_bug.cgi?id=756675
We don't have any way of knowing what the intended size of a XWayland
cursor is supposed to be, so lets do what we do with regular XWayland
surfaces and don't scale them. The result is that cursor sprites of
HiDPI aware X11 clients will show correctly, but non-aware clients may
have tiny cursor sprites.
https://bugzilla.gnome.org/show_bug.cgi?id=755099
Each keyboard focus change ends up calling the MetaWaylandDataDevice
counterpart, we don't need though to notify the current selection
again. In order to fix this, keep track of the current client, and
only emit the relevant signals when the focus switches to another
client.
The situations where wl_data_device.selection were emitted during
focus changes between surfaces of the same client was inocuous most
of the times, although it's prone to inducing confusing behavior
on context menu clipboard actions, as the closing menu triggers a
focus change, which triggers a whole new wl_data_offer being created
and given on wl_data_device.selection, at a time where there's already
ongoing requests on the previous data offer.
https://bugzilla.gnome.org/show_bug.cgi?id=754357
If the transfer is cancelled, the X11SelectionData will be cleared from
the MetaSelectionBridge, although x11_data_write_cb() was invariably
expecting it to be non-NULL.
If the write was cancelled, all the actions done in x11_data_write_cb()
are moot, so just return early. If there's other errors happening
(eg. "connection closed" if the target client happens to crash), we
should still attempt at clearing the data anyway.
https://bugzilla.gnome.org/show_bug.cgi?id=754357
When committing a toplevel surface we might no longer have a MetaWindow
associated with it. The reason may vary but some are: a popup was
dismissed, the client attached and committed a NULL buffer to a
wl_surface with the wl_shell_surface role, the client committed a
buffer to a wl_surface which previously had an toplevel window role
which extension object was destroyed.
https://bugzilla.gnome.org/show_bug.cgi?id=755490
If the drag dest surface suddenly disappears, we may find ourselves
processing an XdndPosition message that was sent before the X11 drag
source had an opportunity to find out.
In that case mutter does know, so double check before processing the
messages.
We try to translate the atom with its corresponding mimetype both back
and forth, which actually breaks if the X11 client chose to announce the
mimetype atom. To do the translation properly, keep track on whether the
source announced the UTF8_STRING atom, and reply back with this only if
that happened.
If the wayland surface isn't available yet when we process the
WL_SURFACE_ID ClientMessage, we schedule a later function to try the
association again after we get a chance to process wayland requests.
This works fine except on cases where the MetaWindow already had a
previous surface attached (i.e. when the xwindow is reparented) since
we only break the existing association on the later function which
means that when processing the old surface's destruction we destroy
the MetaWindow and cancel the pending later function leaving us
without a MetaWindow and an invisible surface.
Fix this by detaching the old surface as soon as possible so that the
MetaWindow survives.
https://bugzilla.gnome.org/show_bug.cgi?id=743339
The saved rect is used to restore a saved window size. We need to
update this when the window is moved to a monitor with different scale,
so that if we unmaximize a window which was moved to a different
monitor while maximized (for example when unplugged) will restore to
the correct size.
https://bugzilla.gnome.org/show_bug.cgi?id=755097
When a window is moved across monitors with different scales, its
rectangle is scaled accordingly. We also need to scale the
unconstrained_rect rectangle, so that moving a window via
meta_window_move_resize() which uses the unconstrained_rect.
https://bugzilla.gnome.org/show_bug.cgi?id=755097
This commits refactors cursor handling code and plugs in logic so that
cursor sprites changes appearance as it moves across the screen.
Renderers are adapted to handle the necessary functionality.
The logic for changing the cursor sprite appearance is done outside of
MetaCursorSprite, and actually where depends on what type of cursor it
is. In mutter we now have two types of cursors that may have their
appearance changed:
- Themed cursors (aka root cursors)
- wl_surface cursors
Themed cursors are created by MetaScreen and when created, when
applicable(*), it will extend the cursor via connecting to a signal
which is emitted everytime the cursor is moved. The signal handler will
calculate the expected scale given the monitor it is on and reload the
theme in a correct size when needed.
wl_surface cursors are created when a wl_surface is assigned the
"cursor" role, i.e. when a client calls wl_pointer.set_cursor. A
cursor role object is created which is connected to the cursor object
by the position signal, and will set a correct texture scale given what
monitor the cursor is on and what scale the wl_surface's active buffer
is in. It will also push new buffers to the same to the cursor object
when new ones are committed to the surface.
This commit also makes texture loading lazy, since the renderer doesn't
calculate a rectangle when the cursor position changes.
The native backend is refactored to be triple-buffered; see the comment
in meta-cursor-renderer-native.c for further explanations.
* when we are running as a Wayland compositor
https://bugzilla.gnome.org/show_bug.cgi?id=744932
Make a surface roles into objects with vfuncs for things where there
before was a big switch statement. The declaration and definition
boilerplate is hidden behind C macros.
https://bugzilla.gnome.org/show_bug.cgi?id=744932
If a surface doesn't have a role, the compositor will not know how, if
or when it will be painted. By adding it to the compositor frame
callback list, the compositor will respond to the client that the
surface has been drawn already which might not be true.
Instead, queue the frame callback in a list that is then processed when
the surface gets a role assigned. The compositor may then, given the
role the surface got, queue the frame callback accordingly.
https://bugzilla.gnome.org/show_bug.cgi?id=744932
Being a "XWayland window" should be considered equivalent to a role,
even though it is not part of any protocol anywhere. The commit doesn't
have any functional difference, but just makes it clear that an
wl_surface managed by XWayland have the same type of special casing as
surface roles as defined by the Wayland protocol.
As the semantics are more explicit given the role is defined, a comment
explaining why the semantics need to be how they are was added.
https://bugzilla.gnome.org/show_bug.cgi?id=744932
Use a better name, use GNOME conventions for error handling, open code the
client error reporting and send the error to the correct resource.
wl_subcompositor doesn't have a role error yet, so continue use some
other error. The only effect of this is error received in the client will
be a bit confusing, it will still be disconnected.
https://bugzilla.gnome.org/show_bug.cgi?id=754215
Older GCC only allows "for (int i" in explicit c99 mode - there's probably
no reason that we can't enable that, but avoiding the construct for
a fast fix.
If the user Alt-Tabs out of the window, we will be left thinking
the Alt key is still pressed since we don't see a release for it.
Solve this and other related issues for the nested X11 compositor
by selecting for KeymapStateMask which causes a KeymapNotify event
to be sent after each FocusIn, and when we get these events, update
the internal XKB state and send any necessary modifiers events to
clients.
https://bugzilla.gnome.org/show_bug.cgi?id=753948
When a client binds an incompatible version, we should terminate it.
This check should only be there for the unstable version, as once it is
declared stable and renamed, future versions will be backward compatible.
https://bugzilla.gnome.org/show_bug.cgi?id=753855
meta_wayland_pointer_get_client_pointer() may be called when the
MetaWaylandPointer as been already shut down, so the hash table will be
NULL at that moment.
The global wl_pointer_gestures object is now created, effectively
bridging pinch/swipe gestures with clients, so they're now
accessible to clients implementing the protocol.
Instead of moving around all the bound pointer resources for a client
when changing focus, keep all the resources bound by a client in a per
client struct, and track the focus by having a pointer to the current
active pointer client struct instance.
This will simplify having wl_pointer extensinos sharing the pointer
focus of the wl_pointer by only having to add them to the pointer
client.
https://bugzilla.gnome.org/show_bug.cgi?id=744104
The spec says:
"A server should avoid signalling the frame callbacks if the surface is not
visible in any way, e.g. the surface is off-screen, or completely obscured
by other opaque surfaces."
We actually do have the information to do that but we are always calling
the frame callbacks in after_stage_paint. So fix that to only call when
when the surface gets drawn on screen.
https://bugzilla.gnome.org/show_bug.cgi?id=739163
If we can't put up a popup because grabbing the pointer fails we
immediately dismiss the popup but the client might have made requests
already, in particular it might have commited the surface and in that
case we should ignore it.
https://bugzilla.gnome.org/show_bug.cgi?id=753237
When a client sets an input region or a opaque region to NULL, it
should still be considered a change to the corresponding region on the
actor. This patch makes sure this state is properly forwarded.
https://bugzilla.gnome.org/show_bug.cgi?id=753222
When placing a popup and the legacy transient wl_shell_surface surfaces,
take the current scale of the window into account. This commit doesn't
fix relative positioning in case a window scale would change, but since
the use case for relative positioning is mostly popups, which would be
dismissed before the parent window would be moved, it should not be that
much of a problem.
https://bugzilla.gnome.org/show_bug.cgi?id=744934
Make meta_wayland_surface_get_toplevel_window return the top most window
in case its a chain of popups. This is to make all popups in a chain
including the top most surface have the same scale.
The reason for this is that popups are mostly integrated part of the
user interface of its parent (such as menus). Having them in a different
scale would look awkward.
Note that this doesn't affect non-popup windows with parent-child
relationship, because such windows are typically not an integral part of
the user interface (settings window, dialogs, ..) and can typically be
moved independently. It would probably make sense to make attached modal
dialogs have the same scale as their parent windows, but modal dialogs
are currently not supported for Wayland clients.
https://bugzilla.gnome.org/show_bug.cgi?id=744934
Since we scale surface actors given what main output their toplevel
window is on, also scale the window geometry coordinates and sizes
(window->rect size and window->custom_frame_extents.top/left) in order
to make the window geometry represent what is being rendered on the
stage.
https://bugzilla.gnome.org/show_bug.cgi?id=744934
Tracking back from the monitor to the output every time we need to
figure out the scale of a window on a monitor is inconvenient, so
propagate the scale from the output to the monitor it is associated
with.
https://bugzilla.gnome.org/show_bug.cgi?id=744934
A MetaWaylandSurface was casted into a ClutterActor, but it should have
been the MetaSurfaceActor.
Move out parent_actor and surface_actor out of the loop while at it
since they won't change when iterating.
https://bugzilla.gnome.org/show_bug.cgi?id=745655
Keep the active position state in its original coordinate space, and
synchronize the surface actor with it when it changes and when
synchronizing the rest of the surface state, in case the surface scale
had changed.
https://bugzilla.gnome.org/show_bug.cgi?id=745655
Previously a MetaWaylandOutput could be removed from the current outputs
table (by being unplugged for example). This would result in the global
object being removed and the MetaWaylandOutput instance freed, but the
wl_resource destructor would still try to remove itself from the list of
resources. Trying to do this, it'd try to access its user data pointer
which would point to the freed MetaWaylandOutput instance, and as a
result crash when trying to manipulate the freed data.
https://bugzilla.gnome.org/show_bug.cgi?id=744453
Whenever a MetaSurfaceActor is painted, update the list of what outputs
the surface is being drawed upon. Since we do this on paint, we
effectively avoids this whenever the surface is not drawn, for example
being minimized, on a non-active workspace, or simply outside of the
damage region of a frame.
DND icons and cursors are not affected by this patch, since they are not
drawn as MetaSurfaceActors. If a MetaSurfaceActor or a parent is cloned,
then we'll check the position of the original actor again when the clone is
drawn, which is slightly expensive, but harmless. If the MetaShapedTexture
instead is cloned, as GNOME Shell does in many cases, then these clones
will not cause duplicate position checks.
https://bugzilla.gnome.org/show_bug.cgi?id=744453
We may access it during painting even if it has been freed. For now,
manually unset it during the MetaWaylandSurface cleanup; in the future
make MetaWaylandSurface a GObject and make the surface pointer a weak
reference.
https://bugzilla.gnome.org/show_bug.cgi?id=744453
Lets use the role when doing role specific commit actions. The
conditions effectively do that anyway, and this way we will get a
compiler warning here whenever we add a new role, as well as we avoid
having different variants of role-determination checks in different
places.
https://bugzilla.gnome.org/show_bug.cgi?id=744453
Firstly, this patch makes MetawaylandDataSource a GObject. This is in
order to easier track its lifetime without adding destroy signals etc. It
also makes the vfunc table GObject class functions instead while at it,
as well as moves protocol specific part of the source into their own
implementations.
An important part of this patch is the change of ownership. Prior to this
patch, MetaWaylandDataDevice would kind of own the source, but for
Wayland sources it would remove it if the corresponding wl_resource was
destroyed. For XWayland clients it would own it completely, and only
remove it if the source was replaced.
This patch changes so that the protocol implementation owns the source.
For Wayland sources, the wl_resource owns the source, and the
MetaWaylandDataDevice sets a weak reference (so in other words, no
semantical changes really). For XWayland sources, the source is owned by
the selection bridge, and not removed until replaced or if the client
goes away.
Given the changes in ownership, data offers may now properly track the
lifetime of a source it represents. Prior to this patch, if an offer with
an XWayland source would loose its source, it wouldn't get notified and
have an invalid pointer it would potentally crash on. For Wayland
sources, an offer would have a weak reference and clean itself up if the
source went away. This patch changes so the behavior is consistent,
meaning a weak reference is added to the source GObject so that the offer
can behave correctly both for Wayland sources and XWayland sources.
https://bugzilla.gnome.org/show_bug.cgi?id=750680
When a possible drag dest client crashes during DnD, it may happen
we receive first the destroy notification for the data_device, and
later the notification for the focus surface. When this happens we
unset the drag_focus_data_device first, and later on
meta_wayland_drag_grab_set_focus(grab, NULL) we assume it still
exists when sending the leave event, leading to mutter crashing
right after.
So, as we don't receive any ordering guarantees about resource
destruction, just prepare the meta_wayland_drag_grab_set_focus()
paths for this.
If a client unsets a selection (calls set_selection with the offer
NULL), this should cause the compositor not to continue sending the
previously set offer.
https://bugzilla.gnome.org/show_bug.cgi?id=750007
Mutter generates a motion event for every button and scroll events,
which confuses Xwayland apps that rely on XMotionEvents for various
purposes, e.g. it fools rxvt jumpy mouse detection code.
Remove the call to notify_motion() from the button and scroll event
handlers to avoid these spurious motion events.
https://bugzilla.gnome.org/show_bug.cgi?id=748705
According to the xdg-shell protocol specification the (x, y) coordinates
passed when creating a popup surface is relative to top left corner of
the parent surface, but prior to this patch, if the parent surface was
a xdg_surface, it'd position it relative to top left corner of the
window geometry of that xdg_surface.
https://bugzilla.gnome.org/show_bug.cgi?id=749716
When DnD is started from an X11 client, mutter now sets up an special
grab that 1) Ensures the drag source keeps receiving events, and 2)
Moves an internal X Window over wayland clients as soon as the pointer
enters over these.
That window will act as the X-side peer for the currently focused
wayland client, and will transform XdndEnter/Position/Leave/Drop
messages into wayland actions. If DnD happens between X11 clients,
the window will be moved away and unmapped, to let these operate as
usual.
https://bugzilla.gnome.org/show_bug.cgi?id=738312
X11 client windows now hook a X11-specific MetaWaylandDragDestFuncs
that converts these into Xdnd* messages, and an additional selection
bridge has been added to take care of XdndSelection, and the data
transfers done through it.
https://bugzilla.gnome.org/show_bug.cgi?id=738312
This will be useful in order to interact with drag dest surfaces in
its windowing-specific ways, although everything defaults to the
wayland vfuncs.
https://bugzilla.gnome.org/show_bug.cgi?id=738312
This piece of code hooks in both wl_data_device and the relevant X
selection events, an X11 Window is set up so it can act as the clipboard
owner when any wayland client owns the selection, reacting to
SelectionRequest events, and returning the data from the wayland client
FD to any X11 requestor through X properties.
In the opposite direction, SelectionNotify messages are received,
which results in the property contents being converted then written
into the wayland requestor's FD.
This code also takes care of the handling incremental transfers through
the INCR property type, reading/writing data chunk by chunk.
https://bugzilla.gnome.org/show_bug.cgi?id=738312
Expose it partly (in internal headers anyway), and pass a vtable for the
data source functions, the wayland vfuncs just delegate operations on the
wl_data_source resource. The resource has been also made optional, although
it'll be present on all data sources from wayland clients.
The ownership/lifetime of the DnD data source has also changed a bit,
belonging now to the MetaWaylandDataDevice like the selection one does, as
we can't guarantee how long it will be needed after the grab is finished,
it will be left inert and replaced the next time DnD is started at worst.
This allows the creation of custom/proxy data sources, which will turn out
useful for X11 selection interoperation.
https://bugzilla.gnome.org/show_bug.cgi?id=738312
The corresponding wl_notify field for destroy_data_device_icon()
is drag_grab->drag_icon_listener, otherwise we're fetching a pointer
that's slightly off where we want.
The stop function currently manually constructs the lock
filename from the display number and also calls unlink
on the same, already known lock filename from the manager
struct.
This commit gets rid fo the manual construction in favor
of the saved lock filename.
https://bugzilla.gnome.org/show_bug.cgi?id=748380
The start function has a few exit paths that need to
perform clean up of the lock file.
This commit consolidates those exit paths at the end
using an out label and gotos.
https://bugzilla.gnome.org/show_bug.cgi?id=748380
Add set_modal ond unset_modal to the gtk_surface interface. When a
surface is modal, the compositor can treat it differently from non-modal
dialogs, for example attach it to the parent window if any. There is
currently no changes to input device focus; it is up to the client to
ignore events to the parent surface that is wanted.
This bumps the gtk_shell version to 2.
https://bugzilla.gnome.org/show_bug.cgi?id=745720
When a parent of a subsurface gets it state applied (either by a
wl_surface.commit, wl_subsurface.set_desync or a recursive
wl_surface.commit on a parent surface), the pending position state
of the subsurface should be applied. If the subsurface is in effective
synchronized mode (i.e. if its in explicit synchronized mode or any of
its parent surfaces is a subsurface in explicit synchronized mode), the
cached state should also be applied at this point, including its
subsurface children, recursively.
https://bugzilla.gnome.org/show_bug.cgi?id=743617
Otherwise we'll access freed memory in the handlers.
The wayland keyboard is released when the seat loses the keyboard
capability which happens when leaving the VT so if there are keymap
changes while switched away from the VT we would crash.
https://bugzilla.gnome.org/show_bug.cgi?id=747263
since commit 8c16ac47c1, we started
creating the login screen on display 1024 instead of display 0.
This defeats this logic in try_display:
display++;
/* If display is above 50, then something's wrong. Just
* abort in this case. */
if (display > 50)
In practice it doesn't matter much since we only have one login
screen in most setups, but we should still fix the bug.
This commit introduces a separate counter to keep try of 50 tries,
rather than assuming "display number == number of tries".
https://bugzilla.gnome.org/show_bug.cgi?id=746545
To avoid integer overflow when scaling "infinite" regions (0, 0)
(INT32_MAX, INT32_MAX), intersect with the surface rect before scaling,
instead of intersecting with the buffer rect afterwards.
https://bugzilla.gnome.org/show_bug.cgi?id=746510
The original code in Weston that this was ported from returned an errno,
not a boolean, so we were inadvertently returning TRUE here during an
error path. Fix that up.
When a client wants to start initialized it my set the maximized state
before having attached any buffers. Before we'd not notify the client of
the new expected size if the previous size was 0x0 as it would normally
mean we'd resize to 1x1, but since this is not always the case, only
avoid notifying the client if the previous size was 0x0 and the result
is 1x1.
https://bugzilla.gnome.org/show_bug.cgi?id=745303
If the wl_surface resource happens to be destroyed before any other
role resource, the destructor for the latter will attempt to
access/modify random memory.
Fix this by ensuring the associated resources are destroyed on the
wl_surface destructor, this will free all associated memory and
remove the resources ahead of their imminent destruction.
https://bugzilla.gnome.org/show_bug.cgi?id=745734
In 3.16, GDM keeps a login screen running on vt1.
This login screen starts an Xwayland instance.
Since it's the first X server to start, it gets
the prized :0 display number.
This commit works around that problem, for now,
by having GDM's display number start at 1024.
https://bugzilla.gnome.org/show_bug.cgi?id=746295
MetaWaylandFrameCallback has been added a surface field, which is then
checked when destroying the surfaces. This prevents unintended callbacks
to run after a surface has been destroyed.
https://bugzilla.gnome.org/show_bug.cgi?id=745163
In order to switch to the correct surface actor scale given the monitor
the surface is on, without relying on the client committing a new state
given some other side effect, sync the surface actor state when the main
monitor associated with the corresponding window changed.
https://bugzilla.gnome.org/show_bug.cgi?id=744933
Since the surface actor knows more about how it draws itself, instead of
pushing texture state (buffer and scale), input region and opaque region
from MetaWaylandSurface after having transformed into what the surface
actor expects, make the surface actor set its own state given what state
the Wayland surface is in.
https://bugzilla.gnome.org/show_bug.cgi?id=744933
Doing this on manage() allows the common MetaWindow initialization to
do the right thing for popups like setting skip_taskbar and
skip_pager.
In particular this avoids gnome-shell's app tracker to create a new
ShellApp instance for every popup.
https://bugzilla.gnome.org/show_bug.cgi?id=745118
Before commit ac448bd42b the pointer,
keyboard, and touch objects were initialized when the seat was created.
Now they're initialized later, when the clutter device manager finds and
loads them.
This commit makes sure we don't try to access those objects if they
aren't initialized.
https://bugzilla.gnome.org/show_bug.cgi?id=744640
The wayland seat event handlers get sent events that
aren't strictly interesting to them (such as events for
hardware devices the seat doesn't support and events for
virtual devices that the seat needs to ignore).
This commit makes sure all uninteresting events get ignored.
If a client creates an xdg_popup given a parent that is a xdg_popup that
is not the most top one in the grab chain, send the
not_the_topmost_popup error.
Also fail a client who destroys a popup that is not the top most one.
https://bugzilla.gnome.org/show_bug.cgi?id=744452
We'll want to expose popup logic outside of meta-wayland-pointer.c and
one day we'll also probably want to add touch support for popups, so
lets move it to its own file. There are no significant semantical
changes, only refactoring.
https://bugzilla.gnome.org/show_bug.cgi?id=744452
This just exposes the type and the singleton getter necessary to make
it available to introspection. We'll expose more functionality as it
becomes needed.
https://bugzilla.gnome.org/show_bug.cgi?id=743745
The commit 97a69cee5a broke the caching of
the surface state when because the frame_callback_list target state was
overwritten after the content had been moved to it.
This commit fixes it by moving the frame list addition after the copy. We
also need to initialize the list since the plain copy put garbage in it.
https://bugzilla.gnome.org/show_bug.cgi?id=743678
Whenever the compositor takes a grab, we're supposed send leave/enter
events to the current surface, which makes sense, as the compositor
has stolen the pointer from the client.
I forget why I added the special case in the first place, but it's
likely a bug that's since been fixed.
This actually fixes a bug: it prevents the need to double-click on
X11 application titlebars when grabbing them.
The input region currently only gets scaled by the surface
scale while ignoring the output scale, which causes input events to not get
delivered correctly for clients on hidpi screens. So take the output scale
into account when doing so.
https://bugzilla.gnome.org/show_bug.cgi?id=739161
This commit is wrong, it assumes that the scale only applies to the one
set by the client but its not. meta_surface_actor_wayland_scale_texture
also handles the output scale. Revert the commit to fix hidpi for wayland
clients like weston-terminal.
This reverts commit 0364ea9140.
https://bugzilla.gnome.org/show_bug.cgi?id=739161
The key event should be interpreted by clients with the modifier state
as it was before the event itself just as in X11 input events.
Achieving this in wayland is a matter of sending the key event first
and the modifiers after (if needed).
This isn't really specified in the wayland protocol but it matches
weston's behavior and should avoid corner cases in clients.
https://bugzilla.gnome.org/show_bug.cgi?id=738238
This reverts commit 33acb5fea0.
The issue here is that the pointer actor does not actually get reset
when the actor's reactivity changes, so we end up with stale picks after
actors are destroyed.
I have a local patch to Clutter for this, but I don't have time to
submit it upstream, so let's just use the ugly code for now.
This reverts commit e496ed50d6.
This was incorrect. wl_surface_destructor actually does the full repick
-- doing it here is dangerous, because the destroy listeners actually
run *before* the destructor, not after, so the surface is still alive.
We never want to send pressed keys to wayland clients on enter. The
protocol says that we should send them, presumably so that clients can
trigger their own key repeat routine in case they are given focus and
a key is physically pressed.
Unfortunately this causes some clients, in particular Xwayland, to
register key events that they really shouldn't handle, e.g. on an
Alt+Tab keybinding, where Alt is released before Tab, clients would
see Tab being pressed on enter followed by a key release event for
Tab, meaning that Tab would be processed by the client when it really
shouldn't.
Since the use case for the pressed keys array on enter seems weak to
us, we'll just fake that there are no pressed keys instead which
should be spec compliant even if it might not be true.
https://bugzilla.gnome.org/show_bug.cgi?id=727178
It only contained a pointer to a wl_resource, which isn't much of
value. Just replace it with the wl_resource instead. Any future private
data should be handled by our future role system.
The actor is updated on DnD grab motion events, properly notified
when dragging finishes, and destroyed if the client/surface disappear
below its feet.
Keeping track of the surface will be necessary in case it is destroyed
during DnD, and the coordinates will be useful when figuring out the
snap back coordinates.
When grabbing with DND, we need to leave the pointer alone and
under the client's control. The code here was a bit messy before about
when it unset the window cursor -- it did it whenever there was no
current surface after repicking, which is a bit wrong, since it will
fire during a drag grab.
Move the check for this to update_cursor_surface, which is our standard
"sync" API for this, and then call update_cursor_surface after we set
the focus.
During a DND grab, pointer->focus_surface is NULL, since the wl_pointer
doesn't have any focused surface (it's in drag mode). In this case, the
drag interface has control of the focus, and when dragging into a NULL
surface, drag_grab_focus won't get called, properly detaching it from
the previous surface.
Let the interface->focus implementation do the fizzling out.
In the future, we should split out wl_pointer's implementation
(pointer->focus_surface) from the Wayland side of the generic pointer
wrapper (pointer->current) and use our event routing system to determine
or similar whether it should go to wl_pointer or wl_data_device.
Some applications, like totem, create keyboard/pointer objects from the
same client, and expect it to work. We made this work a while ago, but
due to an oversight in the code, we increment the serial on button press
for every resource that we need to send events to.
Since operations like move/resize use the grab serial of the devices to
determine whether the operation is exact, we need to make sure the same
serial goes to all devices.
Restructure the code so that all that's in the resource loop is the
sending of the event -- all the calculation that's needed happens
outside.
This fixes moving / resizing the Totem window not working sometimes.
https://bugzilla.gnome.org/show_bug.cgi?id=736840
The fix in d61dde1 regressed the position of popup windows, since the
size was 0x0 when we wanted to do a sole move. Only fizzle out in the
path where we actually *do* resize.
https://bugzilla.gnome.org/show_bug.cgi?id=736812
We only broadcast input to the focus_resource_list, so we need to make
sure it's put in the proper list on startup.
This fixes input not working for windows when they first appear.
Argh. There's always more stuff to fix with keyboard/pointer. Every
single time I think I've fixed it, more stuff pops up.
GTK+ requests get_xdg_surface before attaching a buffer, and since it
might take a long time for GTK+ to get around to attaching a buffer and
committing it, our idle for MOVE_RESIZE will kick in beforehand.
And our idle will try to resize the 0x0 window that currently exists,
constrain it to 1x1, which will send a configure event of 1x1 to the
window while it boots up, causing it to awkwardly resize to the minimum
size of the window.
Make sure that in this case, our idle doesn't cause any problems, and
that we fizzle out any idles like this.
The "proper" way to do this would be to delay the creation of the
MetaWindow until a surface is committed, but that's difficult for a
variety of reasons, and might cause unintended issues with focus.
The last_sent size is effectively what size we should send in configure
requests where the size hasn't changed. Thus, if an app commits a new
size, we need to make sure we respect it and don't reconfigure it with
a size it wasn't expecting when the state changes.
This fixes apps being sent a configure event with 0, 0 on startup,
which was confusing Clutter into displaying a 0x0 viewport.
Windows can be freed at some point after they are unmanaged - because
there is an effect in progress, because a language binding is holding
a reference. Therefore, we need to clean up the later to associate
the xwayland and wayland windows deterministically in an "unamanaged"
handler.
https://bugzilla.gnome.org/show_bug.cgi?id=736694
g_idle_add() makes no guarantee about when it will be run - if Mutter
is busy drawing and blocking glXSwapBuffers() it could happen only
minutes later. Use meta_later_add (META_LATER_BEFORE_REDRAW) instead -
this will deterministically be run after the Wayland socket is read
from but before the next frame is painted.
https://bugzilla.gnome.org/show_bug.cgi?id=736694
Putting X windows and pointers to MetaWindows into a union had a number of
problems:
- It caused awkward initialization and conditionalization
- There was no way to refer to Wayland windows (represented by
MetaWindow *) in the past, which is necessary for the MetaStackTracker
algorithms
- We never even cleaned up old MetaStackWindow so there could be
records in MetaStackWindow pointing to freed MetaWindow.
Replace MetaStackWindow with a 64-bit "stack ID" which is:
- The XID for X Windows
- a "window stamp" for Wayland windows - window stamps are assigned
for all MetaWindow and are unique across the life of the process.
https://bugzilla.gnome.org/show_bug.cgi?id=736559
Add private functions for the test framework to use to find out the
wayland and x11 display names, so they can set up the environment for
children.
https://bugzilla.gnome.org/show_bug.cgi?id=736505
It's possible for a released pointer to have repick / set_focus on it as
part of sync_input_focus. When the pointer is actually re-init'd, it
will memset 0, which can cause corruption as our destroy listener has
already been added.
Released devices should be idempotent, so just make sure method calls on
them don't have any effect.
A wl_surface may have a wl_subsurface interface, but no buffers attached
yet, even though the geometry calculation code for surfaces/subsurfaces
assumes everything has already a buffer.
Just skip subsurfaces that don't have a buffer, those can't be set
a geometry yet, and right now it's crashing accessing the texture from
the NULL surface->buffer.
https://bugzilla.gnome.org/show_bug.cgi?id=735452
This makes it so that MetaSurfaceActorWayland is effectively just a
wrapper actor around MetaShapedTexture with some extra scaling. I think
the MetaSurfaceActor subclassing was a bad idea -- we really should have
these abstractions in much higher levels in the stack than the
compositor.
It doesn't make sense to update it in the surface actor. It's also
theoretically wrong to update the buffer's texture on surface commit,
too, because it's buffer state, not surface state, but I don't think
there's any place we use a wl_buffer without a wl_surface.
The initialization sequence before was quite icky, initializing Clutter
in a few different places depending on what was going on.
Put that all back into main.c
MetaGrabOp is painful and tedious to work with, because it's a
sequential series of values, meaning we have to use a giant unreadable
switch statement to figure out some basic things about the value.
To solve this, modify the encoding for MetaGrabOp and for the specific
window grab operations so that they're a set of bitflags that we can
easily check.
We've long used a switch statement on the grab operation to determine
where events should go. The issue with MetaGrabOp is that it's a mixture
of a few different things, including event routing, state management,
and the behavior to choose during operations.
This leads to poorly defined event routing and hard-to-follow logic,
since it's sometimes unclear what should point where, and our utility
methods for determining grab operations apart can be poorly named.
To fix this, establish the concept of a "event route", which describes
where events should be routed to.
Refuse to create a touch resource if we don't have the capability
(for misbehaving clients), and don't attempt to use touch data
structures that are not initialized.
This is a terrible hack. We need to figure out a better way to do
interactive resizes.
This fixes weird resizing from the left bugs when using GTK+, which is
really slow at acking configures.
This is an easy way to get into an infinite loop where we're constantly
re-sending stuff to the window. If it worked once, it probably won't
work again.
We assume in meta_window_wayland_move_resize that the next commit that
changes the geometry will always be for our next pending operation, so
if we have a move pending on a resize, the next commit will trigger the
move. This is, of course, fundamentally wrong.
We broke this assumption even more now that we don't fizzle out calls to
meta_window_move_resize_internal and now call it on every commit, which
means that a simple damage and then commit would complete a pending
move.
This was even broken by apps like weston-terminal, which, when clicking
on the maximize button, first redraws the terminal with the maximize
button state back on hover on press, and would only redraw when it got
the configure event with the coordinates.
To track the correct commit to apply the move for, we implement the
ack_configure request and ignore all move/resizes that happen before
that.
Right now, we actually fizzle out the entire move/resize if there's a
future pending configure we're waiting on.
The grabbing state is now checked for both pointer/touch devices
within the seat, and the grab start coordinates returned by
meta_wayland_seat_get_grab_info().
https://bugzilla.gnome.org/show_bug.cgi?id=733631
The capability flags are determined from the device types of the slave devices
that are currently attached. This also happens whenever a device is added or
removed, so the capabilities are kept up to date, and clients know about these.
On VT switch, all slave devices are temporarily removed, so the cascade of
signals will make the seat end up with capabililities=0 while input is suspended.
https://bugzilla.gnome.org/show_bug.cgi?id=733563
Anytime the keymap is changed, either directly, or indirectly through the
keyboard capability being released/initialized, there should be a
notification of the modifiers being changed too.
https://bugzilla.gnome.org/show_bug.cgi?id=733563
Otherwise the focus_surface_listener list element becomes stale, and then
mangled if the devices' data is initialized again, and the memory memset().
https://bugzilla.gnome.org/show_bug.cgi?id=733563
This doesn't match what Weston does. I don't know of any apps that this
fixes (we don't have any apps that even use non-zero dx/dy, I don't
think), but this is part of a cleanup for window geometry.
When frame extents change, we might not update the frame rect, but the
buffer rect still needs to be updated. Split out the check for this to
be independent of the check for the frame rect.
This fixes issues that could happen when the window was maximized while
it was in the top-left corner.
The output_id is more of an opaque identifier for the monitor, based on
its underlying ID from the windowing system. Since we also use the term
"output_id" for the output's index, rename our use of the opaque cookie
"output_id" to "winsys_id".
When we changed the setting of the buffer rect to be inside the moving
code to make sure it was updated in places we were moving directly
without any round-trip needed, I removed a code to set the buffer rect
without remembering that's where the size of it was updated.
Add back the code to update the buffer rect.
This fixes Wayland windows not appearing.
It returns FALSE when button_count is not 0. But grabbing for
move/resize is activated by clicking the button, so this condition
disallows the wayland clients to be moved/resized.
https://bugzilla.gnome.org/show_bug.cgi?id=731237
Rather than calculate it speculatively with the current properties
which may be too new or too out of date, make sure it always fits
with the proper definition. We update it when we update the toplevel
window for X11, and when a Wayland surface is committed with a newly
attached buffer.
There is no way this value will ever be read, because we set the
cursor_surface to NULL, this is set at the same time as cursor_surface,
and it's only read if cursor_surface is non-NULL.
Clutter touch events are translated into events being sent down
the interface resource, with the exception of FRAME/CANCEL events,
which are handled directly via an evdev event filter.
The seat now announces invariably the WL_SEAT_CAPABILITY_TOUCH
capability, this should be eventually updated as devices come and
go.
The creation of MetaWaylandTouchSurface structs is dynamic, attached
to the lifetime of first/last touch on the client surface, and only
if the surface requests the wl_touch interface. MetaWaylandTouchInfo
structs are created to track individual touches, and are locked to
a single MetaWaylandTouchSurface (the implicit grab surface) determined
on CLUTTER_TOUCH_BEGIN.
https://bugzilla.gnome.org/show_bug.cgi?id=724442
Smooth scroll event vectors from clutter have the same dimensions as the
ones from from Xi2, i.e. where 1.0 is 1 discrete scroll step. To scale
these to the coordinate space used by wl_pointer.axis
vertical/horizontal scroll events, multiply the vector by 10.
https://bugzilla.gnome.org/show_bug.cgi?id=729601
The last commit added support for the "appmenu" button in decorations,
but didn't actually implement it. Add a new MetaWindowMenuType parameter
to the show_window_menu () functions and use it to ask the compositor
to display the app menu when the new button is activated.
https://bugzilla.gnome.org/show_bug.cgi?id=730752
The requested_rect is a strange name for it, because it's not actually
the rect that the user or client requested all the time: in the case of
a simple move or a simple resize, we calculate some of the fields
ourselves.
To the MetaWindow subclass implementations, it just means "the rect
before we constrained it", so just use the name unconstrained_rect.
This also makes it match the name of the MetaWindow field.
Realistically, the user rect contains the unconstrained window
rectangle coordinates that we want to be displaying, in case
something in the constraints change.
Rename it to the "unconstrained_rect", and change the code to always
save it, regardless of current state.
When metacity was originally being built, the purpose of the user
rect was a lot less clear. The code only saved it on user actions,
with various other calls to save_user_window_placement() and a force
mechanism sprinkled in to avoid windows being snapped back to odd
places when constraints changed.
This could lead to odd bugs. For instance, if the user uses some
extension which automatically tiles windows and didn't pass
user_action=TRUE, and then the struts changed, the window would be
placed back at the last place a user moved it to, rather than where
the window was tiled to.
The META_IS_USER_ACTION flag is still used in the constraints code
to determine whether we should allow shoving windows offscreen, so
we can't remove it completely, but we should think about splitting
out the constrainment policies it commands for a bit more
fine-grained control.
https://bugzilla.gnome.org/show_bug.cgi?id=726714
The default focus interface uses the button count to determine
whether we should update the pointer focused surface. When releasing
an implicit grab, we need to send the button release events to the
implicitly grabbed surface, so we can't reset the focus surface too
soon. We already explicitly set the focus at the end of implicit
grabs, so counting the buttons after is perfectly fine.
If we send out a configure notify for a window and then have some
other kind of state change, we need to make sure that we continue
to send out that new size, rather than the last size the client
sent us a buffer for.
In particular, a client might give us a 250x250 buffer and then
immediately request fullscreen. We send out a configure for the
monitor size and a state that tells it it's full-screen, but then
it takes focus, and since the client hasn't sent us a buffer for
the new size, we tell it it's fullscreen at 250x250.
Fix this.
If we attach to a MetaWindow that disappears before the idle fires,
we'll notice that we can't associate the window properly again and
try to access data on the MetaWindow struct, which might crash.
Install a weak ref that ties the lifetime of the idle to the lifetime
of the MetaWindow.
It seems every GTK+ app does this for some reason at startup. This
is really unfortunate, since we'll have to create and destroy a new
MetaWindow really quickly.
Scale surfaces based on output scale and the buffer scale set by them.
We pick the scale factor of the monitor there are mostly on.
We only handle native i.e non xwayland / legacy clients yet.
https://bugzilla.gnome.org/show_bug.cgi?id=728902
Advertise the scale factor on the output and transform pointer and damage
events as well as input and opaque regions for clients
that scale up by themselves i.e use set_buffer_scale.
We do not scale any 'legacy' apps yet.
https://bugzilla.gnome.org/show_bug.cgi?id=728902
Ugh. So in the fullscreen case, we need to make sure to specify that
it's a MOVE_ACTION so that we move to the saved position, but we
can't do that in the resizing case since we need to use the resized
rectangle.
The flags are really hurting us here. Perhaps we should make it the
client's responsibility to specify a complete rectangle which we
could resize to; then the weird-o logic would be self-contained in
each front-end.
I'm not convinced this covers all cases, especially when we could have
a dangling weird state pointer, but it fixes our existing two testcases.
Restoring the position in our move_resize_internal implementation
is too late. We need to do it at ack-time, before we hand off the
new position to the constraints code.
For the server-initiated resize case, like unmaximize or some forms
of tiling, we dropped the x/y of the server-assigned rectangle on the
floor, which meant the surface didn't move to where it needed to be in
that case. Now, save it internally, and combine it with the dx/dy passed
in during attaches to figure out where we actually need to be.
Make sure to only use it for when we send out a configure notify. We
should use the passed in rectangle for other scenarios, like a
client-initiated resize.
This fixes incorrect surface placement after unmaximization.
For the server-initiated resize case, like unmaximize or some forms
of tiling, we dropped the x/y of the server-assigned rectangle on the
floor, which meant the surface didn't move to where it needed to be in
that case. Now, save it internally, and combine it with the dx/dy passed
in during attaches to figure out where we actually need to be.
This fixes incorrect surface placement after unmaximization.
Looking at the code paths where is_mouse / is_keyboard are used,
all of them should never be run when dealing with a COMPOSITOR
grab op, since they're filtered out above or the method is just
never run during that time.
It's confusing that COMPOSITOR is in here, and requires us to
be funny with other places in code, so just take it out.
pointer->current needs to always be the surface under the pointer,
even when we have a grab. We do need to make sure we keep the focus
surface the same even when we have a grab, though, so add logic
for that.
In order to correctly fix the issue to make sure we only set the
focused surface to NULL during a grab, but not the current surface,
we need to merge update_current_surface back into repick_for_event
so we have more control over the behavior here.
... not when we do an update.
We only repick when we handle events, not when we update. Perhaps
this is a mistake.
Since update runs before handle_event, this means that when we
drop a grab, update will notice the NULL surface, since we haven't
repicked after the event, and then we'll repick the correct surface.
The end result is that you see a root cursor after a grab ends,
rather than the correct window cursor.
This doesn't fix it, since the current surface becomes NULL when
we start the grab. But it does make the code here more correct when
we fix that bug.
I was talking with other people and they became confused at the
term "double-buffered", since we were also talking about
double-buffering in general, e.g. swapping between two buffers.
Instead, we'll adapt the "pending state" nomenclature that we
already use for the field / variable names.
If we have a focused surface, we need to eat up key events, not
just if we have a non-empty focus resource list. The latter would
happen if we have a focused client but it never called get_keyboard.
The latest Xorg / Xwayland has support for -displayfd being used
in conjunction with an explicit display number. Use that to know
when the X server is ready, rather than UNIX signals, because
they're UNIX signals.
If we're sending out a configure event, we can't immediately move the
window; we need to instead wait to apply the new position when the
client sends a new buffer.
dx/dy should be against the regular window's rect, and need to
be ignored when we're resizing. Instead, we use gravity to anchor
the window's new rectangle when resizing.
Sophisticated clients, like those using ClutterGtk, will have more
than one focused resource per client, as both Clutter and GDK will
ask for a wl_pointer / wl_keyboard. Support this naturally using
the same "hack" as Weston: multiple resource lists, where we move
elements from one to the other.
In order to support multiple pointers for the same client, we're
going to need to kill it.
This will cause crashes for now, but will be fixed by the next
commit.
default_grab_focus tries to add implicit grab semantics where
focus won't take effect if there's a pointer button down. This
is not what we want for popup grabs at all, as it's perfectly
valid to want to drag on a menu while there's a button down.
The idea here is that while we take a WM-side grab, like a compositor
grab or a resizing grab, we need to remove the focus from the Wayland
client.
We make a special exception for CLICKING operations, because these
are really an internal state machine while you're pressing on a button
inside a frame, and in this case, we need to not kill the focus.
Really, it is a special case. When the subsurface is synchronous,
commit changes meaning from being applied immediately to being
queued up for replay later. Handle this explicit special case
with an explicit special case in the code.
This means that in all other paths, we can unconditionally
apply the actor immediately.
Even when it doesn't have a role.
This fixes cursors not quite working right, as they're a "detached"
surface without a role since nobody called set_cursor on them yet.
Instead of using commit_attached_buffer / actor_surface_commit.
We want to kill the return values of these methods because we
really should always be calling them, even if the surface doesn't
have a role.
This is also something that we did upstream. Since we want to
introduce an explicit "xdg_transient" window type for tooltips
and popovers, and since "transient_for" is a confusing dumb
80s term lifted from the ICCCM spec, just rename it.
This was changed upstream a little while ago for C++ compatibility.
It's also the more common term for the operation: you close a window,
you don't delete one. In fact, a delete event might seem like it
would be about resource management instead.
Since we get the ClientMessage after the surface is created, there's
no good way to synchronize the two streams. In this case, what we
need to do is delay the surface commit until after we get the
ClientMessage. Ideally, we'd be using a better surface system overall
where committing the surface didn't depend on what type it is, but
oh well, this is a good short-term hack for now.
This is effectively the same, but since we lose the xserver.xml protocol
in the new XWayland DDX, we have to use SIGUSR1 anyway, so might as well
switch over now.
The make_toplevel / window_unmanaging interface has never made
a lot of sense to me. Replace it with set_window, which does
effectively the same thing.
It's still not perfect in the case of XWayland, but I don't think
XWayland will ever make me happy.
It should be META_TYPE_WAYLAND_STAGE, not META_WAYLAND_TYPE_STAGE.
Well, actually, it *should* be META_TYPE_NATIVE_STAGE, because it's
not related to Wayland at all. But that comes later :)
I'm a bit tired of hearing about this when I launch mutter-wayland
nested. Ideally, this would be part of display server integration,
not GNOME integration, so we could simply not make the call when
nested, but oh well.
Previously, a sequence like this would crash a client:
=> surface.attach(buffer)
=> buffer.destroy()
The correct behavior is to wait until we release the buffer before
destroying it.
=> surface.attach(buffer)
=> surface.attach(buffer2)
<= buffer.release()
=> buffer.destroy()
The protocol upstream says that "the surface contents are undefined"
in a case like this. Personally, I think that this is broken behavior
and no client should ever do it, so I explicitly killed any client
that tried to do this.
But unfortunately, as we're all well aware, XWayland does this.
Rather than wait for XWayland to be fixed, let's just allow this.
Technically, since we always copy SHM buffers into GL textures, we
could release the buffer as soon as the Cogl texture is made.
Since we do this copy, the semantics we apply are that the texture is
"frozen" in time until another newer buffer is attached. For simple
clients that simply abort on exit and don't wait for the buffer event
anyhow, this has the added bonus that we'll get nice destroy animations.
If we have a CLICKING grab op we still need to send events to xwayland
so that we get them back for gtk+ to process thus we can't steer
wayland input focus away from it.
https://bugzilla.gnome.org/show_bug.cgi?id=726123
meta_wayland_seat_repick() can be called in various cases while mutter
has a GRAB_OP ongoing which means we could be sending wrong pointer
enter/leave events.
https://bugzilla.gnome.org/show_bug.cgi?id=726123
In particular we need to know about all key events to keep the xkb
state reliable even if the event is then consumed by a global shortcut
or grab and never reaches any wayland client.
We also need to keep track of all pressed keys at all times so that we
can send an updated set or pressed keys to the focused client when a
grab ends.
https://bugzilla.gnome.org/show_bug.cgi?id=722847
Any given clutter event carries the modifier state as it was before it
occured but, for the wayland modifiers event, we want the state
including the current event.
To fix this, we'll keep our xkb_state instance around instead of the
serialized mods.
https://bugzilla.gnome.org/show_bug.cgi?id=722847
