Now it's not very reasonable to determine the default window width based
on the titlebar geometry, as the titlebar can be shrunk to 1px.
Let's use the fixed value of 100px for simplification.
- add LAB_WINDOW_TYPE_INVALID in place of literal -1
- document more clearly that enum lab_view_criteria is a bitset
- other one-off replacements of integer values/types for consistency
Note: variables of type enum lab_view_criteria are already used
extensively throughout the code to contain combinations of the declared
enum values. I am not introducing any new usage here, just changing the
single uint32_t to be consistent with all the other usages.
I like the new common/edge.h. I don't like how inconsistently we use it.
Current situation:
- enum wlr_edges and wlr_direction are designed to be used as bitset,
and are defined compatibly
- enum lab_edge is *also* designed to be used as bitset, but
incompatible with the others (LEFT/RIGHT come before UP/DOWN)
- we use an inconsistent mix of all three *AND* uint32_t (usually with
the WLR_EDGE constants rather than the LAB_EDGE constants), and
convert between them on an ad-hoc basis, sometimes implicitly
Let's clean this up:
- reorder enum lab_edge to be compatible with the two wlr enums
(check this by static_assert)
- use TOP/BOTTOM naming rather than UP/DOWN (matches wlr_edges)
- add constants for the remaining possible combinations of the 4 edges
- use lab_edge for all internal edge/direction fields, consistently
- add lab_edge_is_cardinal() as a sanity check before casting to
enum wlr_direction, and then eliminate all of direction.c/h
Instead of "enum wlr_edges direction", we now have
"enum lab_edge direction" which is not that much better. At least we
are now clear that we're overloading one enum with two meanings.
In 943f5751, I initialized heap-allocated `view_query` used for
`If` actions with `decoration=LAB_SSD_MODE_INVALID`, but I forgot to do
that for stack-allocated `view_query` used for window rules.
When implementing single-axis maximize some time ago, I made the
simplifying assumption that a view couldn't be resized while maximized
(even in only one axis). And indeed for compositor-initiated resize,
we always unmaximize the view first.
However, I didn't account for the client resizing the non-maximized
axis, which we can't (and shouldn't) prevent. When this happens, we
should also update the natural geometry of that single axis so that we
don't undo the resize when un-maximizing.
P.S. xdg-shell clients resizing the *maximized* axis is still an
unsolved problem, exacerbated by the fact that xdg-shell protocol
doesn't allow clients to even know about single-axis maximize.
P.P.S. the view_invalidate_last_layout_geometry() logic may need
similar updates, I'm not sure.
Currently, the dependencies between foreign-toplevel[-internal],
ext-foreign, and wlr-foreign are cyclical and a bit complex.
I suggest we reorganize it into a simpler hierarchy:
foreign-toplevel/
-> foreign.c/h
-> (depends on) ext-foreign.c/h
-> (depends on) wlr-foreign.c/h
The refactored code is smaller and (IMO) easier to follow.
In detail:
- Add include/foreign-toplevel folder mirroring src/foreign-toplevel
- Split foreign-toplevel-internal.h to ext-foreign.h and wlr-foreign.h
- Eliminate ext-/wlr-foreign.c -> foreign.c reverse dependencies
(including internal signals and foreign_request* functions)
- Make struct foreign_toplevel private to foreign.c
Lightly tested with qmpanel (which uses wlr-foreign-toplevel).
v2: reorder foreign-toplevel internal API funcs
If they are not empty, then we are headed for use-after-free shortly.
An assert() failure is easier to debug than UAF, so let's fail early.
Inspired by:
8f56f7ca43
In addition to <snapping><range>, <snapping><cornerRange> configures the
distance from the screen corner to trigger quater window snapping.
Also, new values "up-left", "up-right", "down-left" and "down-right" are
allowed for <action name="(Toggle)SnapToEdge" direction="[value]"> and
<query tiled="[value]">.
This is a common practice in C projects, which simply enforces that
each header must compile cleanly without implicit dependencies on
other headers (see also the previous commit).
In 9e3785f8cd, a heuristic was added to assume that NORMAL and DIALOG
window types were always focusable. (This was before we had the "offer
focus" mechanism in place.)
However, we should still call wlr_xwayland_surface_offer_focus() for
these views, in case they actually don't want focus. (This is uncommon
but has recently been seen with WeChat popups, which have both NORMAL
and UTILITY type.)
To make this possible, refine view_wants_focus() to return either
LIKELY or UNLIKELY for Globally Active input windows. This decouples
the question of "should we try to focus this view" from the actual
mechanism used to do so.
This patch also changes the semantics of scaled_icon_buffer: rather than
calling scaled_icon_buffer_set_app_id() every time an app_id is set, we
can now call scaled_icon_buffer_set_view() just once so that multiple
scaled_icon_buffers bound to a window are automatically updated when an
app_id is set or new icon is set via xdg-toplevel-icon-v1.
Offer focus by sending WM_TAKE_FOCUS to a client window supporting it.
The client may accept or ignore the offer. If it accepts, the surface will
emit a focus_in signal notifying the compositor that it has received focus.
The compositor should then call wlr_xwayland_surface_activate(surface, true).
This is a more compatible method of giving focus to windows using the
Globally Active input model (see wlr_xwayland_icccm_input_model()) than
calling wlr_xwayland_surface_activate() unconditionally, since there is no
reliable way to know in advance whether these windows want to be focused.
v2: add caution not to use view_offer_focus() directly
v3: remove obsolete comment
Ref: 1133bc15ceb2c2bcb6df692acda6bfa39a292ab5
("Transparently restack xwayland surfaces")
In addition, MR 4772 makes sure the hidden windows are stacked at the
bottom, just like what we did with XWAYLAND_VIEW_HIDDEN.
The view->impl functions do not directly support mapping a view while
minimized. Instead, mark it as not minimized, map it, and then minimize
it again.
Fixes: #2627
Before this commit, when a normal window is raised, xwayland thought it's
above always-on-top (AOT) windows even though it's actually below AOT
windows in the scene. This means mouse scroll events may be unexpectedly
sent to normal windows below AOT windows even when the cursor is hovering
over a AOT window.
So this commit fixes it by notifying the correct stacking order (where AOT
windows are placed above normal windows) to xwayland every time the
stacking order is updated.
Other benefits of this commit are:
- It makes the code more readable and predictable by aggregating logic
about stacking order management in xwayland (e.g. shaded windows or
windows in other workspaces should be notified to xwayland as being
placed at the bottom).
- As server->last_raised_view is removed in the previous commit, we were
notifying the stacking order to xwayland every time a window with dialog
windows is clicked (not when clicking a topmost window without dialogs,
due to some optimization in wlroots). This commit fixes this by caching
the window stacking order in xwayland_view->stacking_order and notifying
it to xwayland only when it's updated.
We were skipping in view_move_to_front() if the raised view is already
cached as server->last_raised_view. But this was prone to bugs that
windows cannot be raised even though they are actually not placed at the
top in the scene.
This happened when a window is mapped but view_move_to_front() is not
called in view_impl_map() for some reason. Example cases were:
- a window is minimized before being mapped (#2627)
- a window is mapped while window switcher is active (#2629)
Also, this problem was the root cause of #1640 and #2582, though they are
already fixed.
Therefore, this commit removes server->last_raised_view. In order to
eliminate unnecessary communications between labwc and xwayland (ref:
db591d1), I'll introduce another caching mechanism in the next commit.
This commit moves the check against server->input_mode from the callers
of desktop_focus_view() into desktop_focus_view() itself. This
eliminates code duplications and makes it harder to mess up the window
stacking order while window switching.
I also added the same check in view_minimize() so that minimize requests
from panels never messes up the window stacking order (I think only this
should be described in the release note).
...when app_id is NULL.
Make sure view_get_string_prop() never returns NULL because it is so easy
to misuse. Same for the respective xwayland/xdg impl methods in case
anyone decides to (incorrectly) call them directly in future.
Fixes: #2453
The previous revert fixed the problem of stuck modifier keys with
keybinds in Blender, but made Firefox show its menu bar with Alt-*
keybinds. This is fundamentally inevitable due to the limitation of
wayland protocol, but at least for the default Alt-Tab keybind for
window switcher, we can mitigate this problem by clearing the keyboard
focus when the window switcher is activated. This is what KWin does, and
we decided to follow that.
So in this commit, keyboard and pointer focus are temporarily cleared
while Move/Resize, window switcher and menu interactions and restored
after them. We slightly deviate from KWin as KWin doesn't clear the
keyboard focus while Move/Resize, but it solves our existing problem
that Firefox shows its menu bar after dragging it with default Alt-Drag
mousebind, and this is what Mutter does.
We considered other solutions, but they don't work well:
1. Send wl_keyboard.{leave,enter} every time keybinds/mousebinds are
triggered. This solves the Firefox's menu bar problem, but that
sounds like a workaround and sending unnecessary events every time is
not desirable.
2. Send release events for both modifiers and keys even when they are
bound to keybinds. This is what Mutter is doing, but it looks like an
implementation issue and violates wayland protocol.
