Commit Graph

6176 Commits

Author SHA1 Message Date
Marek Chalupa
17462c21e8 pointer/keyboard: fix setting focus
Set focus to NULL after using the variable
2014-02-19 10:24:36 -05:00
Jasper St. Pierre
c964ef4e01 window-actor: Fix build
I'm bad.
2014-02-19 01:11:42 -05:00
Jasper St. Pierre
0dccc440b6 window-actor: Kill off a nonsensical edge case
We can never have a window actor that represents either the X root
window or the stage window, so it doesn't make sense to bail out
early in case we do.

I'd imagine that this came from a much earlier version of the code
where the compositor was much separate and had its own MapNotify
handling.
2014-02-18 23:36:46 -05:00
Jasper St. Pierre
48f7232492 pointer: Don't send modifiers on mouse enter
I talked to Kristian about this. It should be enough to simply
send it when we activate the surface and give it keyboard focus.
2014-02-18 23:21:06 -05:00
Jasper St. Pierre
ff5867e4d3 pointer: Make the code here a bit clearer 2014-02-18 23:21:06 -05:00
Jasper St. Pierre
a5d950f453 pointer/keyboard: Fix focus setting once again
Yet another large-scale restructuring... this is some messy code.
2014-02-18 23:21:06 -05:00
Jasper St. Pierre
7615d17293 wayland: Tie activate / deactivate to appears-focused
... rather than actual focus. This makes things behave better.
2014-02-18 23:21:06 -05:00
Jasper St. Pierre
374e30043b wayland: Remove width / height from MetaWaylandBuffer
They're in the texture.
2014-02-18 22:43:24 -05:00
Jasper St. Pierre
f771bb88d6 Revert "wayland: Explicitly destroy the surface actor"
This reverts commit 283a81eac0.

This can't be done yet, as it will crash when we try to do a destroy
effect from a plugin. The surface actor needs to outlive the surface
in this case.

Though, the unparenting happening is wrong anyway for a destroy effect.
We need to figure out a sane way of doing this unparenting only after
all effects have finished.
2014-02-18 22:23:08 -05:00
Jasper St. Pierre
7ef8d21e48 wayland: Allow destroying the wl_surface before the xdg_surface
As resource destruction can happen in any order at shutdown, we
need to be flexible here. A client disconnecting without cleaning
up all its resources should not assert fail.
2014-02-18 22:21:33 -05:00
Jasper St. Pierre
c251eb8ec0 window-group: Don't special-case the unredirected window
Since the unredirected window MetaWindowActor is stacked on top, it
will naturally get culled out of the process, so we can remove the
special casing here. Unfortunately, with the way that the code is
currently structured, it's too difficult to actually prevent setting
the clip / visible regions if the window is redirected, so just let
those be set for unredirected windows for now.
2014-02-18 21:29:50 -05:00
Jasper St. Pierre
1e6b3faa83 Fix the input region not working properly
The input region was set on the shaped texture, but the shaped texture
was never picked properly, as it was never set to be reactive. Move the
pick implementation and reactivity to the MetaSurfaceActor, and update
the code everywhere else to expect a MetaSurfaceActor.
2014-02-18 21:29:23 -05:00
Jasper St. Pierre
e62fe956fd window: Enable pinging on Wayland windows
I implemented pinging, but never actually enabled the feature
properly on Wayland surfaces by setting the net_wm_ping hint to
TRUE, causing the fallback path to always be hit.

Rename net_wm_ping to can_ping so it doesn't take on an
implementation-specific meaning, and set it for all Wayland windows.
2014-02-18 20:30:52 -05:00
Jasper St. Pierre
24c5290d7f main: Kill a runtime warning
g_setenv doesn't take NULL for a value.
2014-02-18 20:30:36 -05:00
Jasper St. Pierre
00c8d3c897 xdg-shell: Update to latest renames for focused_set / focused_unset 2014-02-18 19:02:17 -05:00
Jasper St. Pierre
cc13f8f65e wayland: Fix compile error
I tested this with meta_window_get_input_rect and decided to change
it at the last minute. Turns out meta_window_get_rect is unlike all
the others. Just access window->rect directly instead.
2014-02-18 19:01:50 -05:00
Jasper St. Pierre
47f4c9db7b wayland-surface: Correct xdg_popup placement
Now that we call set_custom_frame_extents, the frame rect corresponds
to the "visible window geometry" used for constrainment, while the
x/y fields in get_xdg_popup instead are relative to the surface.
2014-02-18 18:55:54 -05:00
Jasper St. Pierre
4c621cc30f pointer/keyboard: Make sure to clean up the destroy listeners when releasing 2014-02-18 18:39:19 -05:00
Jasper St. Pierre
756a412436 pointer: Make nested grabs work
Something noticed on code inspection. If we have a popup grab,
it will always return FALSE. The code here clearly meant to continue
if we had an existing popup grab from an existing client.
2014-02-18 18:06:55 -05:00
Jasper St. Pierre
be16c2fe71 pointer: Reindent 2014-02-18 18:06:17 -05:00
Jasper St. Pierre
a364c2a96b pointer/keyboard: Make sure not to get stale on client resources as well
Both the pointer/keyboard resource and surface resource can be destroyed
at any point in the destruction process, so we need to have destroy
listeners on both. To make the code easier to follow, rename ->focus
to ->focus_surface at the same time, and rearrange the code so that
the two of them are always grouped together.
2014-02-18 18:03:31 -05:00
Jasper St. Pierre
414259a7f8 wayland: Fix cast warning
It's sometimes hard to see around all the deprecation warnings. Sorry
about that...
2014-02-18 17:26:03 -05:00
Jasper St. Pierre
283a81eac0 wayland: Explicitly destroy the surface actor
This shouldn't change anything right now, but it's better to be
careful than anything else. This is where the actor should be
destroyed.
2014-02-18 17:07:36 -05:00
Jasper St. Pierre
2a145262c7 wayland: Rework how surface destruction works
To prevent the MetaSurfaceActor from being destroyed, we normally
unparent it before we unmanage the window. However, this doesn't
work for XWayland windows, which we unmanage when we get UnmapNotify
or DestroyNotify, not when we get the wl_surface_destroy.

To solve this, add an early hook in meta_window_unmanage that
unparents the surface actor if we have one. At the same time, clean
up the destruction code to remove old comments and assumptions about
how wl_shell behaves.
2014-02-18 17:01:33 -05:00
Jasper St. Pierre
57728b4322 Kill HAVE_WAYLAND 2014-02-17 21:50:32 -05:00
Jasper St. Pierre
ef24fb6296 wayland: Don't use WESTON_LAUNCHER_SOCK
We can simply check the windowing backend.
2014-02-17 21:50:32 -05:00
Jasper St. Pierre
86f057a712 monitor: Add a Wayland backend
We shouldn't crash when trying to run nested under Wayland. Just
use the dummy backend for now until we are able to run with a
system compositor.
2014-02-17 21:50:32 -05:00
Jasper St. Pierre
dcd628d289 wayland: Kill meta_wayland_compositor_is_native
It's unused. Squash the remaining functionality, setting the DRM FD,
into the main path.
2014-02-17 21:15:31 -05:00
Jasper St. Pierre
6038877c4c cursor-tracker: Kill off the last user of is_native 2014-02-17 21:11:57 -05:00
Jasper St. Pierre
4e6321c239 monitor: Kill off another use of is_native 2014-02-17 20:51:35 -05:00
Jasper St. Pierre
304a525744 wayland: Remove is_native from seat
We can easily do this with a Clutter backend check instead...
2014-02-17 20:49:04 -05:00
Jasper St. Pierre
f26de405dd main: Remove clutter / cogl option groups
Neither of these groups are too useful, and all the switches useful
for debugging can also be controlled by envvars.
2014-02-17 20:41:36 -05:00
Jasper St. Pierre
ed18580118 default: Improve the destroy effect as well 2014-02-17 19:27:33 -05:00
Jasper St. Pierre
b65649186d pointer/keyboard: Put the destroy listener on the right resource
We care about when the surface is destroyed, not when the wl_pointer
slash wl_keyboard resource is destroyed.
2014-02-17 19:22:30 -05:00
Jasper St. Pierre
515dc08a97 pointer/keyboard: Do a series of renames to our listener as well 2014-02-17 19:22:30 -05:00
Jasper St. Pierre
2e7a56a28f wayland: Rename resource_destroy_cb to wl_surface_destructor
So it doesn't seem like a wl_listener callback.
2014-02-17 19:22:30 -05:00
Jasper St. Pierre
c485637a61 default: Don't use anchor points
It doesn't work now that we set the pivot point. This breaks the
maximize effect, but it fixes the destroy effect. The maximize effect
looks bad anyway, so it's not too important to me.
2014-02-17 17:57:45 -05:00
Jasper St. Pierre
f9f2a82e18 xwayland: Don't -retro
This means we see black instead of checkerboard for now when resizing,
but that's better.
2014-02-17 17:57:40 -05:00
Jasper St. Pierre
2f4563132a default: Use a consistent rand() for monitor backgrounds
g_random_int() is seeded with /dev/urandom, so it's not consistent.
2014-02-17 14:02:13 -05:00
Jasper St. Pierre
e3a0f2c546 default: Don't wait to show the stage
We can show it immediately now...
2014-02-17 13:57:35 -05:00
Jasper St. Pierre
d39baeb8ad wayland-surface: Destroy the right thing
data is the wl_client, which we don't want to destroy. Destroy
our XdgShell pointer instead.
2014-02-17 12:53:56 -05:00
Jasper St. Pierre
f27f6aab78 Update to new xdg-shell pinging standards 2014-02-16 10:21:22 -05:00
Jasper St. Pierre
a66060e21a Revert "Move window pings to MetaWindow"
This was a bad idea, as ping/pong has moved to a client-specific
request/event pair, rather than a surface-specific one. Revert
the changes we made here and correct the code to make up for it.

This reverts commit aa3643cdde.
2014-02-16 10:21:22 -05:00
Jasper St. Pierre
11aa3c030b wayland-surface: Don't require we manually bump the version for xdg-shell
It's in the protocol as an enum.
2014-02-16 10:21:22 -05:00
Jasper St. Pierre
bd1bec5617 wayland-surface: Prefix xdg-shell methods with "xdg_shell_" 2014-02-16 10:21:22 -05:00
Adel Gadllah
d9659d4b36 wayland-surface: Don't crash when someone tries to run a native app using old gtk
This ends up calling set_dbus_properties with a NULL
window, instead of segfaulting the compositor return so that the broken
app dies instead.

https://bugzilla.gnome.org/show_bug.cgi?id=724472
2014-02-16 16:19:23 +01:00
Ryan Lortie
d043d9943b idle-monitor: avoid XSyncBadAlarm X error
If we fail to find the IDLETIME counter, then the alarm variable will be
uninitialised.  Most code paths are careful to check this before
submitting XSync calls, but there is one check missing.

https://bugzilla.gnome.org/show_bug.cgi?id=724364
2014-02-15 13:14:01 +01:00
Jasper St. Pierre
a0ef7c7142 Move position-changed / size-changed signals to the MetaWindow
They fit more appropriately over here...

https://bugzilla.gnome.org/show_bug.cgi?id=720631
2014-02-14 23:44:39 -05:00
Jasper St. Pierre
4efe4483fb compositor: Delay meta_compositor_add_window until the first show
In order for the compositor to properly determine whether a client
is an X11 client or not, we need to wait until XWayland calls
set_window_id to mark the surface as an XWayland client. To prevent
the compositor from getting tripped up over this, make sure that
the window has been fully initialized by the time we call
meta_compositor_add_window.

https://bugzilla.gnome.org/show_bug.cgi?id=720631
2014-02-14 23:44:38 -05:00
Jasper St. Pierre
aec3edb1cc Always map the client and frame windows
Traditionally, WMs unmap windows when minimizing them, and map them
when restoring them or wanting to show them for other reasons, like
upon creation.

However, as metacity morphed into mutter, we optionally chose to keep
windows mapped for the lifetime of the window under the user option
"live-window-previews", which makes the code keep windows mapped so it
can show window preview for minimized windows in other places, like
Alt-Tab and Expose.

I removed this preference two years ago mechanically, by removing all
the if statements, but never went through and cleaned up the code so
that windows are simply mapped for the lifetime of the window -- the
"architecture" of the old code that maps and unmaps on show/hide was
still there.

Remove this now.

The one case we still need to be careful of is shaded windows, in which
we do still unmap the client window. In the future, we might want to
show previews of shaded windows in the overview and Alt-Tab. In that
we'd also keep shaded windows mapped, and could remove all unmap logic,
but we'd need a more complex method of showing the shaded titlebar, such
as using a different actor.

At the same time, simplify the compositor interface by removing
meta_compositor_window_[un]mapped API, and instead adding/removing the
window on-demand.

https://bugzilla.gnome.org/show_bug.cgi?id=720631
2014-02-14 23:19:41 -05:00