This introduces a new view_impl function: is_transient_for. Similar to
container_has_ancestor but works using the surface parents rather than
the tree.
This patch modifies view_is_visible, container_at and so on to allow
transient views to function normally when they're in front of a
fullscreen view.
Sway sets a default status_command which runs date every second. This
patch removes this behaviour so the user can have a NULL status bar if
desired.
I had to swap swaybar's event_loop_poll and wl_display_flush so that it
would map the initial surface.
This patch makes it so when you run reload, the actual reloading is
deferred to the next time the event loop becomes idle. This avoids
several use-after-frees and removes the workarounds we have to avoid
them.
When you run reload, we validate the config before creating the idle
event. This is so the reload command will still return an error if there
are validation errors. To allow this, load_main_config has been adjusted
so it doesn't apply the config if validating is true rather than
applying it unconditionally.
This also fixes a memory leak in the reload command where if the config
failed to load, the bar_ids list would not be freed.
The previous behaviour was to damage the entire view, which would
recurse into each popup. This patch makes it damage only the popup's
surface, and respect the surface damage given by the client.
This adds listeners to the popup's map and unmap events rather than
doing the damage in the create and destroy functions. To get the popup's
position relative to the view, a new child_impl function get_root_coords
has been introduced, which traverses up the parents.
* Create a view on workspace 1
* Switch to workspace 2 (on the same output) and create a floating
sticky view
* Use criteria to focus the view on workspace 1
Previously, we only moved the sticky containers when using
workspace_switch, but the above method of focusing doesn't call it. This
patch relocates the sticky-moving code into seat_set_focus_warp.
A side effect of this patch is that if you have a sticky container
focused and then switch workspaces, the sticky container will no longer
be focused. It would previously retain focus.
In seat_set_focus_warp, new_output_last_ws was only set when changing
outputs, but now it's always set. This means new_output_last_ws and
last_workspace might point to the same workspace, which means we have to
make sure we don't destroy it twice. It now checks to make sure they're
different, and to make this more obvious I've moved both calls to
workspace_consider_destroy to be next to each other.
container_flatten removes the container from the tree (via
container_replace) before destroying it. When destroying, the recent
changes to handle_seat_node_destroy incorrectly assumes that the
container has a parent.
This adds a check for destroying a container which is no longer in the
tree. If this is the case, focus does not need to be changed.
* Click and hold a scrollbar
* Drag the cursor onto another surface
* While still holding the original button, press and release another
cursor button
* Things get weird
There's two ways to fix this. Either cancel the seat operation and do
the other click, or continue the seat operation and ignore the other
click. I opted for the latter (ignoring the click) because it's easier
to implement, and I suspect a second click during a seat operation is
probably unintentional anyway.
* Have multiple outputs
* Launch swaylock
* Unplug an output (possibly has to be the last "connected" one)
* The swaylock surface on the remaining output would not respond to key
events
This was happening because when the output destroys, focus was not given
to the other swaylock surface.
This patch makes focus be transferred to another surface owned by the
same Wayland client, but only if input was inhibited by the surface
being destroyed, and only if it's in the overlay layer. I figure it's
best to be overly specific and relax the requirements later if needed.
This patch removes a check in seat_set_focus_surface which was
preventing focus from being passed from a layer surface to any other
surface. I don't know of a use case for this check, but it's possible
that this change could produce issues.
Re-focus on the container on which the cursor hovers over. A
special case is, if there are menus or other subsurfaces open
in the focused container. It will prefer the focused container
as long as there are subsurfaces.
This commit starts caching the previous node as well as the
previous x/y cursor position. Re-calculating the previous
focused node by looking at the current state of the cursor
position does not work, if the environment changes.
* New configuration option: raise_floating
(From the discussion on https://github.com/i3/i3/issues/2990)
* By default, it still raises the window on focus, otherwise it
will raise the window on click.
To reproduce the problem, create layout
H[view V[view view view-focused]], then switch to another workspace and
have the previously focused view in the vsplit close (eg. using
criteria, or an mpv video finishing). Return to the workspace using
`$mod+<num>` and the entire vsplit would be focused. This happens
because handle_seat_node_destroy would only set a new focus if the
currently focused view or a parent was being destroyed. To fix it, it
needs to set a sibling of the destroying container to focus_inactive
regardless of the current focus, then restore current focus if needed.
This patch changes the function accordingly. Additionally:
* The function now makes an early return if the node being destroyed is
a workspace.
* set_focus has been renamed to needs_new_focus. This variable is true
if the head focus needs to be changed.
Fixes `hide_edge_borders smart` when gaps are in use.
Implements `hide_edge_borders smart_no_gaps` and `smart_borders
on|no_gaps|off`.
Since `smart_borders on` is equivalent to `hide_edge_borders smart`
and `smart_borders no_gaps` is equivalent to `hide_edge_borders
smart_no_gaps`, I opted to just save the last value set for
`hide_edge_borders` and restore that on `smart_borders off`. This
simplifies the conditions for setting the border.
It's better to use DT_RPATH dynamic section of the elf binary to store
the paths of libraries to load instead of overwriting LD_LIBRARY_PATH
for the whole environment, causing surprises. This solution is much more
transparent and perfectly suitable for running contained installations
of wayland/wlroots/sway.
The code unsetting the LD_LIBRARY_PATH/LD_PRELOAD was also deleted as
it's a placebo security at best - we should trust the execution path
that leads us to running sway, and it's way too late to care about those
variables since we already started executing our compositor, thus we
would be compromised anyway.
When the last output is disconnected, output_disable is called like
usual and evacuates the output to the root->saved_workspaces list. It
then calls root_for_each_container to remove (untrack) the output from
each container's outputs list. However root_for_each_container did not
iterate the saved workspaces, so when the output gets freed the
containers would have a dangling pointer in their outputs list. Upon
reconnect, container_discover_outputs would attempt to use the dangling
pointer, causing a crash.
This makes root_for_each_container check the saved workspaces list,
which fixes the problem.
This changes it back so the path given to swaybg is enclosed in quotes.
Additionally, the only character that is escaped in the path stored is
double quotes now. This makes it so we don't need to keep an exhaustive
list of characters that need to be escaped.
The end user will still need to escape these characters in their config
or when passed to swaybg.
This changes our gaps implementation to behave like i3-gaps.
Our previous implementation allowed you to set gaps on a per container
basis. This isn't supported by i3-gaps and doesn't seem to have a
practical use case. The gaps_outer and gaps_inner properties on
containers are now removed as they just read the gaps_inner from the
workspace.
`gaps inner|outer <px>` no longer changes the gaps for all workspaces.
It only sets defaults for new workspaces.
`gaps inner|outer current|workspace|all set|plus|minus <px>` is now
runtime only, and the workspace option is now removed. `current` now
sets gaps for the current workspace as opposed to the current container.
`workspace <ws> gaps inner|outer <px>` is now implemented. This sets
defaults for a workspace.
This also fixes a bug where changing the layout of a split container
from linear to tabbed would cause gaps to not be applied to it until you
switch to another workspace and back.
When we eventually implement `workspace <ws> gaps inner|outer <px>`,
we'll need to store the gaps settings for workspaces before they're
created. Rather than create a workspace_gaps struct, the approach I'm
taking is to rename workspace_outputs to workspace_configs and then add
gaps settings to that.
I've added a lookup function workspace_find_config. Note that we have a
similar thing for outputs (output_config struct and output_find_config).
Lastly, when freeing config it would create a memory leak by freeing the
list items but not the workspace or output names inside them. This has
been rectified using a free_workspace_config function.
view_is_visible would return false, which meant the view wouldn't
receive a frame done event. view_is_visible needs to make an exception
for floating containers.
This also moves the workspace_is_visible check to an earlier location
for performance reasons.
This does the following:
* Removes the xdg-decoration surface_commit listener. I was under the
impression the client could ignore the server's preference and set
whatever decoration they like using this protocol, but I don't think
that's right.
* Adds a listener for the xdg-decoration request_mode signal. The
protocol states that the server should respond to this with its
preference. We'll always respond with SSD here.
* Makes it so tiled views which use CSD will still have sway decorations
rendered. To do this, using_csd had to be added back to the view struct,
and the border is changed when floating or unfloating a view.
This replaces view.using_csd with a new border mode: B_CSD. This also
removes sway_xdg_shell{_v6}_view.deco_mode and
view->has_client_side_decorations as we can now get these from the
border.
You can use `border toggle` to cycle through the modes including CSD, or
use `border csd` to set it directly. The client must support the
xdg-decoration protocol, and the only client I know of that does is the
example in wlroots.
If the client switches from SSD to CSD without us expecting it (via the
server-decoration protocol), we stash the previous border type into
view.saved_border so we can restore it if the client returns to SSD. I
haven't found a way to test this though.
This commit replaces the non-standard SOCK_NONBLOCK and SOCK_CLOEXEC
flags with two fcntl calls. This makes the file POSIX 2001 compliant,
thus it is no longer necessary to conditionally define, or use internal
(__BSD_VISIBLE) feature test macros.
When a view unmaps, we start a transaction to destroy the container,
then when the transaction completes we destroy the container and unset
the view's container pointer. But if the view has remapped in the
meantime, the view's container pointer will be pointing to a different
container which should not be cleared.
This adds a check to make sure the view is still pointing to the
container being destroyed before clearing the pointer. The freeing of
the title format is also removed as it is already freed when the view
destroys in view_destroy.
If the output being disconnected contains views, and the views are being
relocated to another output of a different size, a transaction must
occur to reconfigure them. This means by the time
container_discover_outputs is called, the output is already disabled and
wlr_output is NULL.
I considered making it check output->wlr_output, but output->enabled
should work just as well and is more descriptive.
This fixes the following. Create these layouts and run move right:
(Initial layout -> expected result -> actual result)
* `H[S[unfocused focused] unfocused]` ->
`H[S[unfocused] focused unfocused]` ->
`H[H[S[unfocused] focused] unfocused]`
* `H[S[unfocused focused] V[unfocused]]` ->
`H[S[unfocused] V[unfocused focused]]` ->
`H[H[S[unfocused] focused] V[unfocused]]`
move_out_of_tabs_stacks was originally made to allow views to move out
of the tabbed/stacked container in the parallel direction, but at some
point this has started working using the regular logic.
If you have swaybar docked to the top, and you create a floating sticky
container and switch workspaces on the same output, the sticky container
would move down by the height of swaybar on each switch.
This happens because when creating the workspace we set the dimensions
to the same as the output, then the subsequent arrange corrects it.
During this arrange, floating containers are translated so they stay
relative to the workspace. This translation needs to not occur for the
initial arrange.
This patch makes workspaces have a zero width and height when first
created, so we can detect whether this is the initial arrange and avoid
translating the floating containers if so.
This adds a `con` argument to `execute_command` which allows you to
specify the container to execute the command on. In most cases it leaves
it as `NULL` which makes it use the focused node. We only set it when
executing `for_window` criteria such as when a view maps. This means we
don't send unnecessary IPC focus events, and fixes a crash when the
criteria command is `move scratchpad` (because we can't give focus to a
hidden scratchpad container).
Each of the shell map handlers now check to see if the view has a
workspace. It won't have a workspace if criteria has moved it to the
scratchpad.
Fixes#2674.
The cause of the issue was in get_pango_layout. When we call
pango_parse_markup, `text` is the escaped string, and the unescaped
string is then computed and written to `buf`. We were then passing the
unescaped string to pango_layout_set_markup, but this function needs the
escaped string. `buf` is not needed and has been removed.
The other part of this PR refactors escape_markup_text to remove the
dest_length argument and removes the -1 return value on error. It now
assumes that you've allocated dest to the correct length.
Firstly, a change had to be made to the container_at functions. If you
create layout `T[view H[view view]]` and hover the second tab, the
container_at functions would return the focus_inactive child. They now
return the split container itself. To compensate for this,
dispatch_cursor_button has been adjusted to find the focus_inactive
child before focusing it.
The actual implementation of wheel scrolling is pretty straightforward.
This uses handle_cursor_axis, so I took a similar approach to
handle_cursor_button (ie. creating a dispatch_cursor_axis function).
root_for_each_container and root_find_container were using incorrect
logic to determine if a container was hidden in the scratchpad.
Containers will have a NULL parent if they are a direct child of a
workspace. Containers will have a NULL workspace if they are hidden in
the scratchpad.
The incorrect check meant that root_for_each_container would run the
callback on scratchpad containers twice. This meant that executing a
command such as `[class="$something"] scratchpad show` would cause the
command to run twice, resulting in the container being shown and hidden
again which is effectively a no op.
Fixes#2655.
When destroying an idle-inhibiting client, idle_inhibit_v1_check_active can get
called from transaction_progress_queue on a view with a null container.
view_is_visible does not handle a view in this state.
seat_get_active_child is used to get the active tiling child in a few
places, such as outputs getting their active workspace and
tabbed/stacked containers getting their visible child. When a workspace
uses a tabbed or stacked layout and contains a focused floating view,
calling seat_get_active_child on the workspace would incorrectly return
the floating view. This changes it so it will return the tiling child.
This fixes the following bug:
* Create layout T[view view] then float one of the views
* Attempt to click the tiling view to give it focus - it wouldn't work
because seat_get_active_child would return the floating view
* Create layout T[view view]
* Move the cursor into the title bar area
* Close both views
Sway would crash because container_at_tabbed would attempt to divide by
zero when there are no children.
The children check isn't needed for the stacked function because it
doesn't divide anything by the number of children.
Fixes#2636.
wordexp p is now initialized to {0} to prevent a segfault on wordfree
in the failure case.
File paths with single quotes and double quotes are now supported. The
quote can either be wrapped in the other quote or escaped with three
backslashes.
Additionally to make passing file paths with double quotes to swaybg
easier, instead of enclosing the path given to swaybg in quotes, all
spaces, single quotes, and double quotes in the resulting path are now
escaped with a single backslash.
The assertion can be (rightfully) triggered by creating layout
V[H[view view] view] and moving the top right view to the right.
After removing the assertion I found the container being moved needs its
size reset to prevent it from being sized wrongly after arranging.
Since the `client.{focused,focused_inactive,unfocused,urgent}` commands
change colors, the textures need to be updated otherwise the textures
and the rest of the title bar may utilize different colors.
* Make container_add_sibling's `after` argument a boolean.
* Use a constant for drop layout border
* Make thickness an int
* Add button state check
* Move comments in seat_end_move_tiling
When workspace_wrap_children is called on a workspace which has a
fullscreen child and the fullscreen child is a direct child of the
workspace, sway would crash.
The workspace's fullscreen pointer is unset when the fullscreen
container is detached and applied again when added to a parent, but in
this case the parent hadn't yet been added to the workspace which meant
con->workspace was NULL.
The fix makes container_handle_fullscreen_reparent return if there's no
workspace, and the fullscreen pointer is reapplied in
workspace_wrap_children.
Fixes#2401 (aka #2558)
Previously, when switching windows, pointer focus was not changed until the pointer was moved. This makes the pointer enter happen immediately, without the side effects of other attempted fixes.
Added default values for all nodes, because the i3 get_tree always returns
all fields inside the json objects.
Add geometry and window for views. Window is only availabel on
x11 windows otherwise it's NULL.
Calculate percent only if parent is not empty to avoid division by 0.
This does the following:
* Adds a baseline argument to get_text_size (the baseline is the
distance from the top of the texture to the baseline).
* Stores the baseline in the container when calculating the title
height.
* Takes the baseline into account when calculating the config's max font
height.
* When rendering, pads the textures according to the baseline so they
line up.
There was code that attempted to fill in the gap below the title texture
when the texture isn't tall enough, but this only worked when the output
was positioned at 0,0. The reason is that render_rect expects a box
passed in a hybrid layout-local/output-buffer-local system, and we were
passing purely output-buffer-local. I've added a comment documenting
this.
By the way, we can't use layout-local coordinates for the rectangle box
because in some cases we set the box based on a texture size. Texture
sizes are buffer-local, and we'd have to divide them to bring it back to
layout-local which means losing precision. We could use
output-buffer-local coordinates for the box, but this would require
translating the coordinates from layout-local to output-buffer-local in
many places during rendering.
This patch also vertically centers the text inside the title bar.
This fixes pinentry-gtk-2, but might make other views floating which
would otherwise be tiled. This patch is more of a trial which could end
up becoming a permanent fix.
To reproduce the problem this is fixing, create H[view view view],
fullscreen one of the views and close it. The entire workspace will be
given focus rather than one of the siblings.
This happens because we emit the destroy event, so the seat code tries
to find a new focus, but the view it finds is still believed to be
hidden by the fullscreen view so it's discarded and the workspace is
used instead.
This clears the workspace's fullscreen pointer prior to emitting the
destroy event so that the seat code finds an appropriate new focus.
When rendering, the workspace for the output needs to be retrieved from
the output's `current` state. output_get_active_workspace returns the
pending workspace, which crashes if the pending workspace is new and
hasn't completed a transaction yet.
These are the same as seat_set_focus, but accept a specific type rather
than using nodes. Doing this adds more typesafety and lets us avoid
using &con->node which looks a little ugly.
This fixes a crash that pretty much nobody would ever come across. If
you have a bindsym for "focus" with no arguments and run it from an
empty workspace, sway would crash because it assumes `container` is not
NULL.
* Fullscreen a view
* Run `focus <direction>` where there is no output in that direction
The output returned was rightfully NULL, which needs to be handled.
It was incorrectly determining that the container being moved and the
destination had the same parent, which resulted in tree corruption. Both
parents can be NULL but the containers may belong to different
workspaces.
To reproduce, create layout H[V[view] view] in one workspace then move a
view left or right from another output into that workspace.
Suppose the following:
* Transactions are already in progress - say transaction A.
* View A maps, which creates transaction B and appends it to the
transaction queue.
* View B maps, which creates transaction C and appends it to the queue.
* View A unmaps, which creates transaction D and appends it to the
queue.
* Transaction A completes, so transaction B attempts to save View A's
buffer, but this doesn't exist so it saves nothing.
* Rendering code attempts to render View A, but there is no saved buffer
nor live buffer that it can use.
Rather than implement an elaborate solution for a rare circumstance,
I've take the safe option of just not rendering anything for that view.
It means that if you reproduce the scenario above, you might get the
title and borders rendered but no surface.
Prior to f5b9815128, children of tabbed
and stacked containers would have their container size and position set
to the same as the tabbed/stacked container. Normally this would be a
problem for a layout such as T[V[view]], but there was some code in the
arrange functions which would check if the grandparent of the view was a
tabbed or stacked container and would offset the view's Y accordingly.
Commit f5b9815128 changed the box to
exclude the titlebar for all tabbed/stacked children so that the
grandparent check could be removed. But this meant the title was not
covered in the container and wasn't damaged when the child changed its
title.
This patch changes it so that a child of a tabbed/stacked container will
have its box include the title bar if the child is a view, but not if
it's a layout container. This fixes the title damage issue while
avoiding the grandparent check in the arrange functions, and matches
what we see visually.
seat_execute_command needs to check the flags on `binding_copy`, as
`binding` will be a dangling pointer after a reload command.
handle_keyboard_key needs to set the next_repeat_binding for
non-reloads prior to executing the command in case the binding is
freed by the reload command.
* Was crashing when a view was moved to the scratchpad (prev focus had
no parent).
* Was crashing when a hidden scratchpad view unmaps because it has no
workspace.
When changing focus from a view in one workspace to an empty workspace
using `focus <direction>`, the view in the previous workspace would keep
focused styling. This is because the check to unfocus it was only done
in the container case and not workspace case, so it's been moved out of
both.
This commit changes the meaning of sway_container so that it only refers
to layout containers and view containers. Workspaces, outputs and the
root are no longer known as containers. Instead, root, outputs,
workspaces and containers are all a type of node, and containers come in
two types: layout containers and view containers.
In addition to the above, this implements type safe variables. This
means we use specific types such as sway_output and sway_workspace
instead of generic containers or nodes. However, it's worth noting that
in a few places places (eg. seat focus and transactions) referring to
them in a generic way is unavoidable which is why we still use nodes in
some places.
If you want a TL;DR, look at node.h, as well as the struct definitions
for root, output, workspace and container. Note that sway_output now
contains a workspaces list, and workspaces now contain a tiling and
floating list, and containers now contain a pointer back to the
workspace.
There are now functions for seat_get_focused_workspace and
seat_get_focused_container. The latter will return NULL if a workspace
itself is focused. Most other seat functions like seat_get_focus and
seat_set_focus now accept and return nodes.
In the config->handler_context struct, current_container has been
replaced with three pointers: node, container and workspace. node is the
same as what current_container was, while workspace is the workspace
that the node resides on and container is the actual container, which
may be NULL if a workspace itself is focused.
The global root_container variable has been replaced with one simply
called root, which is a pointer to the sway_root instance.
The way outputs are created, enabled, disabled and destroyed has
changed. Previously we'd wrap the sway_output in a container when it is
enabled, but as we don't have containers any more it needs a different
approach. The output_create and output_destroy functions previously
created/destroyed the container, but now they create/destroy the
sway_output. There is a new function output_disable to disable an output
without destroying it.
Containers have a new view property. If this is populated then the
container is a view container, otherwise it's a layout container. Like
before, this property is immutable for the life of the container.
Containers have both a `sway_container *parent` and
`sway_workspace *workspace`. As we use specific types now, parent cannot
point to a workspace so it'll be NULL for containers which are direct
children of the workspace. The workspace property is set for all
containers, except those which are hidden in the scratchpad as they have
no workspace.
In some cases we need to refer to workspaces in a container-like way.
For example, workspaces have layout and children, but when using
specific types this makes it difficult. Likewise, it's difficult for a
container to get its parent's layout when the parent could be another
container or a workspace. To make it easier, some helper functions have
been created: container_parent_layout and container_get_siblings.
container_remove_child has been renamed to container_detach and
container_replace_child has been renamed to container_replace.
`container_handle_fullscreen_reparent(con, old_parent)` has had the
old_parent removed. We now unfullscreen the workspace when detaching the
container, so this function is simplified and only needs one argument
now.
container_notify_subtree_changed has been renamed to
container_update_representation. This is more descriptive of its
purpose. I also wanted to be able to call it with whatever container was
changed rather than the container's parent, which makes bubbling up to
the workspace easier.
There are now state structs per node thing. ie. sway_output_state,
sway_workspace_state and sway_container_state.
The focus, move and layout commands have been completely refactored to
work with the specific types. I considered making these a separate PR,
but I'd be backporting my changes only to replace them again, and it's
easier just to test everything at once.
Fixes#2568
The binding that gets stored in the keyboard's `repeat_binding` would
get freed on reload, leaving a dangling pointer.
Rather than attempt to unset the keyboard's `repeat_binding` along with
the other bindings, I opted to just not set it for the reload command
because there's no point in reloading repeatedly by holding the binding.
This disables repeat bindings for the reload command.
As we now need to detect whether it's a reload command in two places,
I've added a binding flag to track whether it's a reload or not.
Depends on https://github.com/swaywm/wlroots/pull/1222.
I don't know of a program that sets the state to modal without setting
the window type, but I know the modal property works because logging the
property shows it's true for the Firefox Open File dialog.
May as well make it as easy as possible for users who are coming from
i3.
This also changes the `border` command to accept a thickness when
setting the border to normal. This makes it work the same way as the
`default_border` command. Eg. `border normal 5`
If sway is reloaded using a bindsym which has multiple commands, it
failed to detect the reload command, didn't create a duplicate of the
binding and would crash because the reload command frees the bindings.
For example:
mode system {
bindsym r reload, mode default
}
In this example, the binding->command is "reload, mode default".
Fixes#2545
This moves the arrange_windows call into the arrange_layers function,
where we know the output actually needs to be arranged.
Additionally, we shouldn't set focus to the parent of an unknown
container type, because the parent may be an output and this causes a
crash because outputs can't have direct focus.
Fixes#2543
Rootston calls "wlr_xwayland_destroy" and "wl_display_destroy_clients"
on shutdown, but these were not called by Sway. Without them, Sway
crashes on exit before the display destroy event handler could be
called. This causes two problems:
- The TTY is not reset, and it locks up after exiting Sway.
- drmDropMaster is not called, and the implicit drop (that should
occur when the DRM fd is closed) seems not to be working in some
scenarios (e.g. if you have a tmux session running - maybe the fd
is retained somehow by tmux?). In other words, it you exit Sway,
you can't start it (or any other program that wants to be DRM
master) again until you close all your tmux sessions.
When collecting focus to save into the transaction state, the workspace
needs to look in the tiling list only.
As seat_get_focus_inactive_tiling returns any descendant, the list also
needs to be traversed back up to the direct child of the workspace.
Fixes#2532
When there's multiple transactions in the queue, sway can take a
shortcut by checking if they all operate on the same set of containers.
If they do, it can skip all but the last transaction. The way we tested
for transactions which used the same containers was to exclusive-or
their con IDs together, but this has proved not only to be ineffective
but also has the potential to make sway crash.
This patch replaces the exclusive-or with a loop and container
comparison.
When moving a container to an inactive workspace on a different output, this will change the focus on the destination output back to its last active workspace
* In layout command, arrange parent of parent - not sure why this is
needed but it is
* Remove gap adjustment when rendering
* Workspace should use outer gaps, not inner
* Add exceptions for tabbed and stacked containers
* Don't mess with gap state when splitting a container
This commit changes the arrange code in a way that will support type
safe arguments.
The arrange_output et al functions are now public, however I opted not
to use them directly yet. I've kept the generic arrange_windows there
for convenience until type safety is fully implemented. This means this
patch has much less risk of breaking things as it would otherwise.
To be type safe, arrange_children_of cannot exist in its previous form
because the thing passed to it could be either a workspace or a
container. So it's now renamed to arrange_children and accepts a list_t,
as well as the parent layout and parent's box.
There was some code which checked the grandparent's layout to see if it
was tabbed or stacked and adjusted the Y offset of the grandchild
accordingly. Accessing the grandparent layout isn't easy when using type
safe arguments, and it seemed odd to even need to do this. I determined
that this was needed because a child of a tabbed container would have a
swayc Y matching the top of the tab bar. I've changed this so a child of
a tabbed container will have a swayc Y matching the bottom of the tab
bar, which means we don't need to access the grandparent layout. Some
tweaks to the rendering and autoconfigure code have been made to
implement this, and the container_at code appears to work without
needing any changes.
arrange_children_of (now arrange_children) would check if the parent had
gaps and would copy them to the child, effectively making the
workspace's gaps recurse into all children. We can't do this any more
without passing has_gaps, gaps_inner and gaps_outer as arguments to
arrange_children, so I've changed the add_gaps function to retrieve it
from the workspace directly.
apply_tabbed_or_stacked_layout has been split into two functions, as it
had different logic depending on the layout.
Lastly, arrange.h had an unnecessary include of transaction.h. I've
removed it, which means I've had to add it to several other files.
When we have type safety we'll need to have functions for
workspace_add_tiling and so on. This means the existing container
functions will be just for containers, so they are being moved to
container.c. At this point layout.c doesn't contain much else, so I've
relocated everything and removed the file.
* container_swap and its static functions have been moved to the swap
command and made static.
* container_recursive_resize has been moved to the resize command and
made static.
* The following have been moved to container.c:
* container_handle_fullscreen_reparent
* container_insert_child
* container_add_sibling
* container_add_child
* container_remove_child
* container_replace_child
* container_split
* enum movement_direction and sway_dir_to_wlr have been moved to util.c.
Side note: Several commands included layout.h which then included
root.h. With layout.h gone, root.h has to be included by those commands.
This list includes disabled outputs.
When sway_container is demoted, we'll need to store the root's children
(ie. enabled outputs) in the sway_root. It makes sense to put these in a
list called `outputs`, so I'm renaming the existing list in advance.
* container_move is only called from the move command
* container_move_to was called from both the move command and the sticky
command, but the sticky command can easily not call it
* container_get_in_direction is only called from the focus command
Moving these functions to their respective commands gives better
separation of code and removes bloat from layout.c. These functions will
need to be refactored to take advantage of type safety, so separating
them will make this easier to refactor.
The following static functions have also been moved:
* is_parellel
* invert_movement
* move_offs
* container_limit
* workspace_rejigger
* move_out_of_tabs_stacks
* get_swayc_in_output_direction
They were all used by the move functions, except for the last one which
is used by focus.
Other changes:
* index_child has been renamed to container_sibling_index, moved to
container.c and made public
* sway_output_from_wlr has been renamed to output_from_wlr_output, moved
to output.c and made public
* container_handle_fullscreen_reparent has been made public
* sway_dir_to_wlr has been made public
No changes have been made to any of the moved functions, other than
updating calls to functions that have been renamed.
This changes the destroy functions to the following:
* output_begin_destroy
* output_destroy
* workspace_begin_destroy
* workspace_destroy
* container_begin_destroy
* container_destroy
* view_begin_destroy
* view_destroy
The terminology was `destroy` and `free`, and it has been changed to
`begin_destroy` and `destroy` respectively.
When the last output is disconnected, its workspaces will now be stashed
in the root. Upon connection of a new output they will be restored.
There is a new function `workspace_consider_destroy` which decides
whether the given workspace should be destroyed or not (ie. empty and
not visible).
Calling container_begin_destroy will no longer automatically reap the
parents. In some places we want to reap the parents and in some we
don't, so this is left to the caller.
container_reap_empty_recursive and container_reap_empty have been
combined into one function and it will recurse up the tree.
When a workspace is moved to another output, or the output it's on
changes its global layout position, the floating containers on that
workspace should be translated by the same amount as the workspace. This
keeps the floating containers in the same position relative to the
workspace.
A check is done to make sure the floating container's center point isn't
being moved off screen. If it is, it is centered within the workspace.
Fixes part of #2500.
Improves upon 18e425ed by using the first assigned workspace instead of
the last one. The order isn't explicitly guaranteed to be the same as in
the config, but in general works.
Fixes#2490.
To be honest I'm not sure why this fixes the issue.
I observed that I could only make the view jump if I resized it to the
smallest possible size first. Then I had a suspicion that we were
accidentally factoring in the title and border sizes into the view size
when it uses CSD. So I changed that and it appears to have fixed the
jumping issue.
I guess when we factor the title and borders in, we send a configure to
the surface with a size smaller than the minimum, and it comes back with
a surface at the minimum size. We interpret this as an unexpected
resize, and this somehow makes it jump.
Previously we used a reparent event to detect when a view changes
parent, then sent an output enter/leave to the surfaces if needed. This
worked for tiling views but not floating views, as floating views can
intersect another output without changing parent.
The solution implemented for floating views also applies cleanly to
tiling views, so the previous method has been completely replaced and
the reparent event has been removed.
This introduces a new function container_discover_outputs. This function
compares the container's `current` position to the outputs, sends enter
and leave events as needed, and keeps track of which outputs it's
intersecting in a new `container->outputs` list. If it has entered a new
output with a different scale then the title and marks textures will
also be recreated at the new scale.
The function is called when a transaction applies. This is convenient as
it means we don't have to call it from various places.
There is imperfect rendering when a floating view overlaps two outputs
with different scales. It renders correctly for the most recently
entered output, but there is only one title texture so it renders
incorrectly on the old output.
Fixes#2482
We were removing the saved buffer when one transaction applies, then
didn't have a new buffer to save when the next transaction ran. This
made the rendering code crash as it had no surface to use.
This commit makes it continue to hold the buffer if the view is
destroying and has more transactions. Additionally, a check is added
when saving the buffer to make sure there's no one already there.
Workspaces previously had a magical `workspace->floating` container,
which had a layout of L_FLOATING and whose children were actual floating
views. This allowed some conveniences, but was a hacky solution because
the container has to be exempt from focus, coordinate transactions with
the workspace, and omit emitting IPC events (which we didn't do).
This commit changes it to be a list directly in the sway_workspace. The
L_FLOATING layout is no longer used so this has been removed as well.
* Fixes incorrect check in the swap command (it checked if the
containers had the L_FLOATING layout, but this layout applied to the
magical container).
* Introduces workspace_add_floating
This makes all debug options stored in a single struct rather than in
various places, changes/fixes the behaviour of existing options, and
introduces some new options.
* Fixes damage issues with `-Drender-tree` texture (by removing scissor)
* Offsets the render tree overlay's `y` position for those who have
swaybar at the top
* Replaces `-Ddamage=rerender` with `-Dnodamage`
* Replaces `-Ddamage=highlight` with `-Dhighlight-damage`
* Replaces `-Dtxn-debug` with `-Dtxn-wait`
* Introduces `-Dnoatomic`
* Removes the `create_time` and `ms_arranging` figures from transactions
and the log message. Transactions are created after arranging and the
create time is of no significance.
* Fixes `-Dtxn-debug` (now `-Dtxn-wait`) not working.
This introduces the following `for_each` functions:
* root_for_each_workspace
* root_for_each_container
* output_for_each_workspace
* output_for_each_container
* workspace_for_each_container
And introduces the following `find` functions:
* root_find_output
* root_find_workspace
* root_find_container
* output_find_workspace
* output_find_container
* workspace_find_container
* container_find_child
And removes the following functions:
* container_descendants
* container_for_each_descendant
* container_find
This change is preparing the way for demoting sway_container. Eventually
these functions will accept and return sway_outputs, sway_workspaces and
sway_containers (meaning a C_CONTAINER or C_VIEW).
This change also makes it easy to handle abnormalities like the
workspace floating list, root's scratchpad list and (once implemented)
root's saved workspaces list for when there's no connected outputs.
This commit renames container_sort_workspaces to output_sort_workspaces
and moves it to output.c.
This also renames container_wrap_children to workspace_wrap_children and
moves it to workspace.c. This function is only called with workspaces.
This fixes a race condition flicker when unfloating a view which uses
client side decorations.
When the view is floated it has using_csd = true, so the decorations are
not drawn. When unfloating it it changes to false, but this change
wasn't part of transactions so it could potentially render the
decorations around the view while it's waiting for the transaction to
apply.
Fixes#2467.
This commit introduces seat_get_focus_inactive_floating to supplement
seat_get_focus_inactive_tiling, and uses it during `focus mode_toggle`
which fixes a focus bug.
This also refactors the seat_get_focus_inactive functions so that they
do their selection logic themselves rather than offloading it to
seat_get_focus_by_type which was getting bloated. seat_get_focus_by_type
is now removed.
Lastly, this commit changes seat_get_focus to just return the first
container in the focus stack rather than looping and calling
seat_get_focus_by_type.
The original purpose of this commit is to replace some for loops with
list_find. But while doing this I found the workspace_prev_next_impl
functions to be difficult to read and also contained a bug, so I
refactored them and fixed the bug.
To reproduce the bug:
* Have two outputs, where the left output has workspaces 1, 2, 3 and the
right output has workspaces 4, 5, 6. Make workspace 2 focused_inactive
and workspace 4 focused.
* Run `workspace prev`.
* Previously it would visit the left output, then apply `workspace prev`
to workspace 2, which focuses workspace 1.
* Now it will focus the rightmost workspace on the left output
(workspace 3).
The refactoring I made to the workspace functions are:
* Added the static keyword.
* They now accept an int dir rather than bool, to avoid an unnecessary
conversion.
* Rather than preparing start and end variables for the purpose of
iterating, just iterate everything.
* Replace for loops with list_find.
* Don't call workspace_output_prev_next_impl (this fixes the bug).
Commit 4b8e3a885b makes it so only one
transaction is committed (ie. configures sent) at a time. This commit
removes the now-unnecessary code which was used to support concurrent
committed transactions.
* Instead of containers storing a list of instructions which they've
been sent, it now stores a single instruction.
* Containers now have an ntxnrefs property. Previously we knew how many
references there were by the length of the instruction list.
* Instructions no longer need a ready property. It was used to avoid
marking an instruction ready twice when they were in a list, but this is
now avoided because there is only one instruction and we nullify the
container->instruction pointer when it's ready.
* When a transaction applies, we no longer need to consider releasing
and resaving the surface, as we know there are no other committed
transactions.
* transaction_notify_view_ready has been renamed to
view_notify_view_ready_by_serial to make it consistent with
transaction_notify_view_ready_by_size.
* Out-of-memory checks have been added when creating transactions and
instructions.
This fixes an issue where views might commit to a transaction ahead of
the first one, and applying the first transaction causes us to save a
buffer of the wrong size.
There was a separate function dispatch_cursor_button_floating which
dealt with the resize and move operations, but as resize is not really
limited to floating views, it doesn't make as much sense to have this
separate. So both functions are now combined into one.
Additionally, dispatch_cursor_button now uses a pattern of returning
early instead of using else-ifs.
* The OP_RESIZE seat operation has been renamed to OP_RESIZE_FLOATING,
and OP_RESIZE_TILING has been introduced.
* Similar to the above, seat_begin_resize and handle_resize_motion have
been renamed and tiling variants introduced.
* resize.c's resize_tiled has to be used, so container_resize_tiled has
been introduced in resize.c to allow external code to call it.
This allows for a color to be set when the wallpaper does not fill the
entire output. If specified, the fallback color is also used when the
image path is inaccessible.
Rationale: Sticky containers are always assigned to the visible
workspace.
The basic idea here is to check the destination's output (move.c:190).
But if the command was `move container to workspace x` then a workspace
might have been created for it. We could destroy the workspace in this
case, but that results in unnecessary IPC events.
To avoid this, the logic for `move container to workspace x` has been
adjusted. It now delays creating the workspace until the end, and uses
`workspace_get_initial_output` to determine and check the output before
creating it.
* Removes container_floating_move_to_container, instead opting to put
that logic in container_move_to
* In the seat code, focusing a floating view now updates the pending
state only and lets the next transaction carry it over to the current
state. This is required, otherwise it would crash.
* When unfullscreening a floating container, an output check is now done
to see if it should center it.
In a multi-output setup, if a sticky container is on one output and
focus is on the other output, and you run (eg) `workspace 1` to focus
the workspace containing the sticky container, an infinite loop would
occur. It would loop infinitely because it would remove the sticky
container from the workspace, add it back to the same workspace, and
then decrement the iterator variable.
The fix just wraps the loop in a workspace comparison.
The back_and_forth condition is intended to be handled in the else-if
block, but this was never reached because it remained in the first
block's conditions.
container_move_to handled moving containers to new parents, as well as
moving workspaces to new outputs.
This commit removes the workspace-moving code from this function and
introduces workspace_move_to_output. Moving workspaces using
container_move_to only happened from the move command, so it's been
implemented as a static function in that file.
Simplifying container_move_to makes it easier for me to fix some issues
in #2420.
I've got the following SIGSEGV when terminating sway:
```
Program terminated with signal SIGSEGV, Segmentation fault.
0x00005607dc603af5 in view_unmap (view=0x5607dcb3d350) at ../sway/tree/view.c:599
599 if (surviving_ancestor->type >= C_WORKSPACE) {
```
surviving_ancestor was NULL at that time
This commit is trying to fix this problem.
- Some platforms don't expose kill() unless _POSIX_C_SOURCE is defined.
- fork(), execl(), and setsid() need unistd.h on some platforms.
Basically, this fixes some platform-specific build errors.
This creates a root.c and moves bits and pieces from elsewhere into it.
* layout_init has been renamed to root_create and moved into root.c
* root_destroy has been created and is called on shutdown
* scratchpad code has been moved into root.c, because hidden scratchpad
containers are stored in the root struct
Calling container_at_view fails an assertion if the container isn't a
view. Calling tiling_container_at works correctly, as that function
checks if the container is a view and calls container_at_view if so.
When a view unmaps, normally the surviving ancestor (ie. after reaping)
needs to be arranged. When a fullscreen view unmaps, it arranges the
workspace rather than the surviving ancestor, but didn't handle cases
where the workspace itself was reaped. This happens if the workspace is
not currently shown and the fullscreen view was the last container on
that workspace.
This commit rewrites this part of view_unmap so it's more readable, and
fixes the crash by not arranging the workspace if it's been reaped. Note
that it no longer arranges the output under any circumstance - this
wasn't required anyway.
* seat_set_focus_warp lacked a container NULL check
* view mapping code needs to use seat_get_focus_inactive
Also, seat_set_focus_warp triggered the wrong IPC event if focus was a
workspace, which resulted in swaybar not showing the workspace as
active.
wlroots uses wl_event_loop_add_signal to handle SIGUSR1 from Xwayland.
wl_event_loop_add_signal works by masking the signal and receiving it from a
signalfd. The signal mask is preserved across fork and exec, so subprocesses
spawned by Sway start with SIGUSR1 masked. Most subprocesses do not expect this
and never unmask the signal, resulting in missing functionality or unexpected
behavior for processes that use SIGUSR1 (such as i3status).
Fix this by unmasking all signals between fork and exec.
This fixes two issues which were both introduced in #2396.
First issue:
The PR changes the location of the buffer save to transaction_apply, but
puts it inside the should_configure block. For unmapping (destroying)
views, should_configure returns false so it wasn't saving the buffer. If
a frame was rendered between the unmap and the transaction applying then
it would result in a crash.
Second issue:
If a destroying view is involved in two transactions, we must not
release the buffer between the transactions because there is no live
buffer to grab any more.
When a container is moved from, say, workspace 1 to workspace 2, workspace 2 is focused in order to arrange the windows before focus is moved back to workspace 1, which caused a workspace:focus event from workspace 2 to workspace 1 to be emitted. This commit inhibits that event.
Fixes#2364.
Suppose a view is 600px wide, and we tell it to resize to 601px during a
resize operation. We create a transaction, save the 600px buffer and
send the configure. This buffer is saved into the associated
instruction, and is rendered while we wait for the view to commit a
601px buffer.
Before the view commits the 601px buffer, suppose we tell it to resize
to 602px. The new transaction will also save the buffer, but it's still
the 600px buffer because we haven't received a new one yet.
Then suppose the view commits its original 601px buffer. This completes
the first transaction, so we apply the 601px width to the container.
There's still the second (now only) transaction remaining, so we render
the saved buffer from that. But this is still the 600px buffer, and we
believe it's 601px. Whoops.
The problem here is we can't stack buffers like this. So this commit
removes the saved buffer from the instructions, places it in the view
instead, and re-saves the latest buffer every time the view completes a
transaction and still has further pending transactions.
As saved buffers are now specific to views rather than instructions, the
functions for saving and removing the saved buffer have been moved to
view.c.
The calls to save and restore the buffer have been relocated to more
appropriate functions too, favouring transaction_commit and
transaction_apply rather than transaction_add_container and
transaction_destroy.
Fixes the render and container_at order for popups.
Fixes#2210
For rendering:
* render_view_surfaces has been renamed to render_view_toplevels
* render_view_toplevels now uses output_surface_for_each_surface (which
is now public), as that function uses wlr_surface_for_each_surface which
doesn't descend into popups
* Views now have a for_each_popup iterator, which is used by the
renderer to render the focused view's popups
* When rendering a popup, toplevels (xdg subsurfaces) of that popup are
also rendered
For sending frame done, the logic has been updated to match the
rendering logic:
* send_frame_done_container no longer descends into popups
* for_each_popup is used to send frame done to the focused view's popups
and their child toplevels
For container_at:
* floating_container_at is now static, which means it had to be moved
higher in the file.
* container_at now considers popups for the focused view before checking
containers.
* tiling_container_at has been introduced, so that it doesn't call
container_at recursively (it would check popups recursively if it did)
Now 'repeat_delay' and 'repeat_rate' control the initial delay
and rate (per second) of repeated binding invocations.
If the repeat delay is zero, binding repetition is disabled.
When the repeat rate is zero, the binding is repeated exactly
once, assuming no other key events intervene.
Each sway_keyboard is provided with a wayland timer event source.
When a valid keypress binding has been found, a callback to
handle_keyboard_repeat is set. Any key event will either clear
the callback or (if the new key event is a valid keypress binding)
delay the callback again.
Example config that produces the crash (with a single output):
workspace 1
workspace 2
Prior to this commit, container_workspace_free would manually mark the
L_FLOATING container as destroying and free it. This assumed the
L_FLOATING container would never be involved in a transaction. This was
a safe assumption when it was implemented, but became an incorrect
assumption once parent/child relationships became transactionised.
This commit removes the L_FLOATING free from container_workspace_free.
When the workspace is destroyed, it starts the normal destroy process on
the L_FLOATING container so it can be freed via transactions.
Also fixes a crash when unfloating a window. It needs to add it back to
the tiling tree as a sibling rather than a child, because the reference
container might be a view.
This introduces seat_get_focus_inactive_tiling and updates
`focus mode_toggle` to use it instead, because the previous method
wasn't guaranteed to return a tiling view.
It would focus the split container rather than the child.
This commit makes it track the child and the split container separately
and send the surface click to the child.
Things worth noting:
* When a fullscreen view unmaps, the check to unset fullscreen on the
workspace has been moved out of view_unmap and into container_destroy,
because containers can be fullscreen too
* The calls to `container_reap_empty_recursive(workspace)` have been
removed from `container_set_floating`. That function reaps upwards so it
wouldn't do anything. I'm probably the one who originally added it...
* My fix (b14bd1b0b1) for the tabbed child
crash has a side effect where when you close a floating container, focus
is not given to the tiled container again. I've removed my fix and
removed the call to `send_cursor_motion` from `seat_set_focus_warp`. We
should consider calling it from somewhere earlier in the call stack.
The solution used in 073ac425d5 doesn't
work in all cases because the freed instruction might be ahead in the
list, not necessarily behind.
The new solution delays running the queue until after the loop has
finished iterating, thus avoiding the problem completely.
In set_instructions_ready, calling set_instruction_ready may cause any
number of transactions to get applied, which removes them from the list
being iterated. The iteration variables need to be adjusted
accordingly.
* Move workspace selection into separate function
* Instead of keeping a `prev_focus` variable, do the check in
`should_focus` (ie. views can only take focus if they're mapped into the
active workspace)
* Fix assign-to-output - it previously set `prev_focus` but should be
`target_sibling`
* Remove call to `workspace_switch` as we'll only ever focus the view if
it's in the active workspace
After setting the keymap, try to enable NumLock and disable CapsLock.
This only works if sway has the xkb master state and controls the keyboard.
Prepare configuration settings for later use as well.
when using 2 display, if scaling is different
`container_update_textures_recursive` is called when moving workspace on
different display.
We need to call `container_update_title_textures` only for container of type
"CONTAINER" or "VIEW" in order to be consistent with the assert in
`update_title_texture`.
The crash only occurs if the mouse cursor is above the tabbed container
when the last child is closed.
Introduced in 03d49490cc, over a week ago
and unnoticed until now :O
The above commit changes the behaviour of a focus change. When you
change focus, it sends pointer motion which makes the client set a new
cursor image. We already had this behaviour for workspace switching, but
this commit adds it for view switching too, such as in a tabbed
container or when closing a view.
The sequence of events that leads to the crash is:
* The last child of a tabbed container unmaps, which triggers a
`destroy` event before we've cleaned up the child or reaped the tabbed
container.
* The seat code listens to the `destroy` event and removes the seat
container from the focus stack. As part of this, it decides to set focus
to the parent (my fix alters this decision).
* When setting focus to the new parent, the container motion is sent as
per the previously mentioned commit.
* The motion code uses `container_at`, which encounters the tabbed
container and its child in a half destroyed state, and everything blows
up from there.
`con->parent` is needed because scratchpad containers don't have parents
if they're hidden, so this probably fixes a crash when a hidden
scratchpad container closes too.
The `con->parent->children->length > 1` check should catch any cases
where the parent is about to be reaped.
The rendering code doesn't use the exclusive input surface at all
anymore to decide to skip rendering of shell surfaces. This fixes
a weird situation in which a client requests exclusive input but
isn't an overlay layer surface.
The renderer also renders all overlay surfaces in this situation,
not just one. This simplifies the code and fixes rendering when
there are more than one overlay surfaces (e.g. for a virtual
keyboard to type the lockscreen password).
This makes it so if you hold mod and right click on a surface to resize
it, the resize direction is chosen based on which quarter of the surface
you've clicked. The previous implementation only resized towards the
bottom right.
The mouse binding logic is inspired/copied from the
keyboard binding logic; we store a sorted list of the
currently pressed buttons, and trigger a binding when
the currently pressed (or just recently pressed, in
the case of a release binding) buttons, as well as
modifiers/container region, match those of a given
binding.
As the code to execute a binding is not very keyboard
specific, keyboard_execute_command is renamed to
seat_execute_command and moved to where the other
binding handling functions are. The call to
transaction_commit_dirty has been lifted out.
First, the existing sway_binding structure is given an
enumerated type code. As all flags to bindsym/bindcode
are boolean, a single uint32 is used to hold all flags.
The _BORDER, _CONTENTS, _TITLEBAR flags, when active,
indicate in which part of a container the binding can
trigger; to localize complexity, they do not overlap
with the command line arguments, which center around
_TITLEBAR being set by default.
The keyboard handling code is adjusted for this change,
as is binding_key_compare; note that BINDING_LOCKED
is *not* part of the key portion of the binding.
Next, list of mouse bindings is introduced and cleaned up.
Finally, the binding command parsing code is extended
to handle the case where bindsym is used to describe
a mouse binding rather than a keysym binding; the
difference between the two may be detected as late as
when the first key/button is parsed, or as early as
the first flag. As bindings can have multiple
keycodes/keysyms/buttons, mixed keysym/button sequences
are prohibited.
cursor_set_image only uploads the named image if it doesn't match the
previous named image. This means when setting the cursor image to a
surface as given by a client, we have to clear the currently stored
image.
Implements the following commands:
* move scratchpad
* scratchpad show
* [criteria] scratchpad show
Also fixes these:
* Fix memory leak when executing command with criteria
(use `list_free(views)` instead of `free(views)`)
* Fix crash when running `move to` with no further arguments
This allows to update the title even if the view doesn't commit.
This is useful e.g. when a terminal sets its toplevel title to
the currently running command and when the view isn't visible.
Also does a few other related things:
* Now uses enum wlr_edges instead of our own enum resize_edge
* Now uses wlr_xcursor_get_resize_name and removes our own
find_resize_edge_name
* Renames drag to move for consistency
When interactively resizing some views (eg. Nautilus), new transactions
are added to the queue faster than the client can process them.
Previously, we would wait for the entire queue to be ready before
applying any of them, but in this case the transactions would time out,
giving the client choppy performance.
This changes the queue handling so it applies the transactions up to the
first waiting transaction, without waiting for the entire queue to be
ready.
This implements the following:
* `floating_modifier` configuration directive
* Drag a floating window by its title bar
* Hold mod + drag a floating window from anywhere
* Resize a floating view by dragging the border
* Resize a floating view by holding mod and right clicking anywhere on
the view
* Resize a floating view and keep aspect ratio by holding shift while
resizing using either method
* Mouse cursor turns into resize when hovering floating border or corner
The directive sets the timeout before an urgent view becomes normal
again after switching to it from another workspace.
Also:
* When an xwayland surface removes the urgent hint while the timer is
active, we now ignore the request. This happens as soon as the view
receives focus, so it was effectively making the timer pointless.
* The timeout is now only applied when switching to it from another
workspace.
This removes the urgency stuff from the commit handler and puts it in a
new set_hints handler instead. This allows the xwayland surface to
become urgent without having to commit (which doesn't happen if it's on
an non-visible workspace).
Fixes#2303, as well as a crash.
To replicate the crash:
* Have multiple outputs
* In config: for_window [<criteria>] workspace foo
* Also in config: workspace foo output <left-output-name>
* Focus the right output, and ensure workspace foo doesn't exist
* Launch the app that triggers the criteria
When the view maps, it calls workspace_switch which calls
send_set_focus which calls cursor_send_pointer_motion which calls
transaction_commit_dirty. This call to transaction_commit_dirty is not
meant to happen at this time because the tree isn't guaranteed to be in
a consistent state, but I'm not sure how exactly this leads to the crash
or render issues.
In this case the transaction is already committed by the view
implementation's handle_map function. So the solution is to remove it
from cursor_send_pointer_motion and add it to the other functions in
cursor.c which call cursor_send_pointer_motion.
This allows to send wl_pointer.enter when switching between views
in a split/tabbed layout for instance. This (1) updates the cursor
image accordingly (2) makes it unnecessary to move the mouse before
scrolling. It's harmless to always call cursor_send_pointer_motion
because in case the focused surface hasn't changed this is a no-op.
The `last_focus != NULL` condition is required otherwise
cursor_send_pointer_motion will crash when sway starts up (the
sway_output doesn't yet have a workspace).
When an xwayland view is mapped, the IPC urgent event was being sent on
every surface commit.
I had intentionally ommitted the check because I figured an urgent
surface could update its urgent timestamp by sending urgent a second
time. But that's not how it works in xwayland's case, and it makes for
more complicated code.
Introduces a command to manually set urgency, as well as rendering of
urgent views, sending the IPC event, removing urgency after focused for
one second, and matching urgent views via criteria.
Rather than maintain copies of the entire focus stack, this PR
transactionises the focus by introducing two new properties to the
container state and using those when rendering.
* `bool focused` means this container has actual focus. Only one
container should have this equalling true in its current state.
* `struct sway_container *focus_inactive_child` points to the immediate
child that was most recently focused (eg. for tabbed and stacked
containers).
We currently have several ways of setting debug flags, including command
line arguments, environment variables, and compile-time macros. This
replaces the lot with command line flags.
This PR changes the way we handle transactions to a more simple method.
The new method is to mark containers as dirty from low level code
(eg. arranging, or container_destroy, and eventually seat_set_focus),
then call transaction_commit_dirty which picks up those containers and
runs them through a transaction. The old methods of using transactions
(arrange_and_commit, or creating one manually) are now no longer
possible.
The highest-level code (execute_command and view implementation
handlers) will call transaction_commit_dirty, so most other code just
needs to set containers as dirty. This is done by arranging, but can
also be done by calling container_set_dirty.
Now the scroll_button will not accept:
- letters on string beginning;
- negative numbers.
What is tolerated:
- letters after number;
- rational numbers: the fraction after dot will be omitted.
This commit introduces a scroll_button option, which is intended to be
used with scroll_method. Now user can edit his sway config and add an
scroll_button option to device section.
The title and marks textures would have their height set from the
config's computed max font height, but the textures were not regenerated
when the config's max font height changed which made a gap appear.
Rather than making it regenerate the title textures every time the
config font height was changed, I've changed it to just make the
textures the height of the title itself and fill any gap when rendering.
Also, the title_width and marks_width variables have been renamed to
make it more obvious that they are in output-buffer-local coordinates.
Fixes#1936.
The only user of this function would copy the string right away
to get rid of the const flag anyway, and freeing a const string
afterwards might work but is not meant to be done according to the
json-c API.
If we set an instruction as ready twice, it decreases the transaction's
num_waiting a second time and applies the transaction earlier than it
should. This no doubt has undesired effects, probably resulting in a use
after free.
Hopefully fixes the first part of #2207.
wl_event_source_remove() is illegal after display has been destroyed,
so just destroy everything when we still can.
==20392==ERROR: AddressSanitizer: heap-use-after-free on address 0x607000001240 at pc 0x00000048e86e bp 0x7ffe4b557e00 sp 0x7ffe4b557df0
READ of size 8 at 0x607000001240 thread T0
#0 0x48e86d in wl_list_insert ../common/list.c:149
#1 0x7fdf673d4d7d in wl_event_source_remove src/event-loop.c:487
#2 0x41b742 in ipc_terminate ../sway/ipc-server.c:94
#3 0x40b1ad in main ../sway/main.c:440
#4 0x7fdf6664c18a in __libc_start_main ../csu/libc-start.c:308
#5 0x409359 in _start (/opt/wayland/bin/sway+0x409359)
0x607000001240 is located 48 bytes inside of 72-byte region [0x607000001210,0x607000001258)
freed by thread T0 here:
#0 0x7fdf692c4880 in __interceptor_free (/lib64/libasan.so.5+0xee880)
#1 0x7fdf673d371a in wl_display_destroy src/wayland-server.c:1097
previously allocated by thread T0 here:
#0 0x7fdf692c4c48 in malloc (/lib64/libasan.so.5+0xeec48)
#1 0x7fdf673d4d9e in wl_event_loop_create src/event-loop.c:522
#2 0x40acb2 in main ../sway/main.c:363
#3 0x7fdf6664c18a in __libc_start_main ../csu/libc-start.c:308
When you have an unfocused container (so one view is focused_inactive),
and you focus any other view in that container, the view with
focused_inactive was not damaged. This is because we damaged the
previous focus and new focus, but needed to damage the parent of the new
focus.
We would previously run all config commands without the environment,
which would appear to work as our socket name is the default one, but
wayland clients would start up in the wrong sway session.
(This explains why 'sometimes' my swayidle processes wouldn't die with
sway, as they weren't listening to the correct socket)
Fixes#2192.
seat_get_active_current_child is intended to return a child of the given
container which has finished its mapping transaction and is able to be
rendered on screen. The previous implementation was capable of returning
a pending child, which caused a child of a tabbed or stacked view to be
rendered prematurely while it was mapping.
If init fails halfway through it will call the destroy function,
which needs some coherent stuff filled.
Allocate with calloc and fill in what cannot fail first
Found through static analysis.
The check didn't include && ws_num < 100 so l would always be 1 or 2
Instead of fixing logic it's simpler to just call snprintf twice to get
length and use that.
Also change malloc failure check to sway_assert because both callers of
this function do not do null check and would segfault...
Found through static analysis.
No logic change here, this one is mostly to please static analyzer:
- client->fd can never be -1 (and if it could, close() a few lines below
would have needed the same check)
- we never send permission denied error (dead code)
ipc_send_reply already does client disconnect on error, so we shouldn't
do it again.
We also need to process current index again as disconnect removes client
from the list we currently are processing (this is an indexed "list")
Found through static analysis.
size_t/ssize_t are 8 bytes on 64bit systems, so use the proper size to
transmit that information.
This could lead to ridiculously large alloc as len is not initialized to zero
Found through static analysis
- child would leak in the workspace_record_pid path
- removing malloc lets us get rid of That Comment nobody seems
to remember what it was about
- we would leak pipe fds on first fork failling
- we didn't return an error if second fork failed
- the final executed process still had both pipe fds
(would show up in /proc/23560/fd in launched programs)
- we would write twice to the pipe if execl failed for some reason
(e.g. if /bin/sh doesn't exist?!)
When you spawn a process with the exec command, sway now notes the
workspace you had focused and the pid of the child process, then assigns
that workspace to the child when its window appears.
Some of this is carried over from sway 0.15, but with some major
refactoring and centralization of state.
That event comes from the toplevel and not the surface, so would cause
a use-after-free on destroy if the toplevel got destroyed first:
==5454==ERROR: AddressSanitizer: heap-use-after-free on address 0x6110001ed198 at pc 0x000000472d10 bp 0x7ffc19070a80 sp 0x7ffc19070a70
WRITE of size 8 at 0x6110001ed198 thread T0
#0 0x472d0f in wl_list_remove ../common/list.c:157
#1 0x42e159 in handle_destroy ../sway/desktop/xdg_shell_v6.c:243
#2 0x7fa9e5b28ce8 in wlr_signal_emit_safe ../util/signal.c:29
#3 0x7fa9e5afd6b1 in destroy_xdg_surface_v6 ../types/xdg_shell_v6/wlr_xdg_surface_v6.c:101
#4 0x7fa9e5d98025 in destroy_resource src/wayland-server.c:688
#5 0x7fa9e5d98091 in wl_resource_destroy src/wayland-server.c:705
#6 0x7fa9e27f103d in ffi_call_unix64 (/lib64/libffi.so.6+0x603d)
#7 0x7fa9e27f09fe in ffi_call (/lib64/libffi.so.6+0x59fe)
#8 0x7fa9e5d9bf2c (/lib64/libwayland-server.so.0+0xbf2c)
#9 0x7fa9e5d983de in wl_client_connection_data src/wayland-server.c:420
#10 0x7fa9e5d99f01 in wl_event_loop_dispatch src/event-loop.c:641
#11 0x7fa9e5d98601 in wl_display_run src/wayland-server.c:1260
#12 0x40a2f4 in main ../sway/main.c:433
#13 0x7fa9e527318a in __libc_start_main ../csu/libc-start.c:308
#14 0x40b749 in _start (/opt/wayland/bin/sway+0x40b749)
0x6110001ed198 is located 152 bytes inside of 240-byte region [0x6110001ed100,0x6110001ed1f0)
freed by thread T0 here:
#0 0x7fa9e7c89880 in __interceptor_free (/lib64/libasan.so.5+0xee880)
#1 0x7fa9e5affce9 in destroy_xdg_toplevel_v6 ../types/xdg_shell_v6/wlr_xdg_toplevel_v6.c:23
#2 0x7fa9e5d98025 in destroy_resource src/wayland-server.c:688
previously allocated by thread T0 here:
#0 0x7fa9e7c89e50 in calloc (/lib64/libasan.so.5+0xeee50)
#1 0x7fa9e5b00eea in create_xdg_toplevel_v6 ../types/xdg_shell_v6/wlr_xdg_toplevel_v6.c:427
#2 0x7fa9e27f103d in ffi_call_unix64 (/lib64/libffi.so.6+0x603d)
The toplevel only notifies the compositor on destroy if it was mapped,
so only listen to events at map time.
A flash of background was happening for two reasons:
1) We were using the xsurface's dimensions to check if the surface is
ready, but these are pending dimensions.
2) In my particular setup, the default geometry of the xsurface does not
intersect any output, which prevented it from receiving a frame done
event. This made the transaction time out and the client would only
redraw once it's been rendered.
We were arranging a parent which may have been deleted by the reaper,
which meant the `current` children list of the surviving parent had a
dangling pointer.
Instead, we now reap the workspace.
The view was configured with the container coordinates.
Although they were right on the first configure, they
changed after a XCB_CONFIGURE_REQUEST, when the
border was already drawn.
Rather than allocate a structure and expect callers to free it, take a
pointer to an existing struct as an argument.
This function is no longer called anywhere though.
- fixes a double-free error when access() failed.
- refactor code to make memory managment (alloc/free) more straightforward
- do not bring the temporary wordexp_t struct around
- do not postpone errors handling
Both sway_output and sway_layer_shell listen to wlr's output destroy event,
but sway_layer_shell needs to access into sway_output's data strucure and needs
to be destroyed first.
Resolve this by making sway_layer_shell listen to a new event that happens at
start of sway_output's destroy handler
Fixes this kind of use-after-free:
==1795==ERROR: AddressSanitizer: heap-use-after-free on address 0x612000191ef0 at pc 0x00000048c388 bp 0x7ffe308f0410 sp 0x7ffe308f0400
WRITE of size 8 at 0x612000191ef0 thread T0
#0 0x48c387 in wl_list_remove ../common/list.c:157
#1 0x42196b in handle_destroy ../sway/desktop/layer_shell.c:275
#2 0x7f55cc2549fa in wlr_signal_emit_safe ../util/signal.c:29
#3 0x7f55cc22cf68 in layer_surface_destroy ../types/wlr_layer_shell.c:182
#4 0x7f55cc22d084 in layer_surface_resource_destroy ../types/wlr_layer_shell.c:196
#5 0x7f55cc4ca025 in destroy_resource src/wayland-server.c:688
#6 0x7f55cc4ca091 in wl_resource_destroy src/wayland-server.c:705
#7 0x7f55cc22c3a2 in resource_handle_destroy ../types/wlr_layer_shell.c:18
#8 0x7f55c8ef103d in ffi_call_unix64 (/lib64/libffi.so.6+0x603d)
#9 0x7f55c8ef09fe in ffi_call (/lib64/libffi.so.6+0x59fe)
#10 0x7f55cc4cdf2c (/lib64/libwayland-server.so.0+0xbf2c)
#11 0x7f55cc4ca3de in wl_client_connection_data src/wayland-server.c:420
#12 0x7f55cc4cbf01 in wl_event_loop_dispatch src/event-loop.c:641
#13 0x7f55cc4ca601 in wl_display_run src/wayland-server.c:1260
#14 0x40bb1e in server_run ../sway/server.c:141
#15 0x40ab2f in main ../sway/main.c:432
#16 0x7f55cb97318a in __libc_start_main ../csu/libc-start.c:308
#17 0x408d29 in _start (/opt/wayland/bin/sway+0x408d29)
0x612000191ef0 is located 48 bytes inside of 312-byte region [0x612000191ec0,0x612000191ff8)
freed by thread T0 here:
#0 0x7f55ce3bb880 in __interceptor_free (/lib64/libasan.so.5+0xee880)
#1 0x42f1db in handle_destroy ../sway/desktop/output.c:1275
#2 0x7f55cc2549fa in wlr_signal_emit_safe ../util/signal.c:29
#3 0x7f55cc23b4c2 in wlr_output_destroy ../types/wlr_output.c:284
#4 0x7f55cc1ddc20 in xdg_toplevel_handle_close ../backend/wayland/output.c:235
#5 0x7f55c8ef103d in ffi_call_unix64 (/lib64/libffi.so.6+0x603d)
previously allocated by thread T0 here:
#0 0x7f55ce3bbe50 in calloc (/lib64/libasan.so.5+0xeee50)
#1 0x42f401 in handle_new_output ../sway/desktop/output.c:1308
#2 0x7f55cc2549fa in wlr_signal_emit_safe ../util/signal.c:29
#3 0x7f55cc1d6cbf in new_output_reemit ../backend/multi/backend.c:113
#4 0x7f55cc2549fa in wlr_signal_emit_safe ../util/signal.c:29
#5 0x7f55cc1deac7 in wlr_wl_output_create ../backend/wayland/output.c:327
#6 0x7f55cc1db353 in backend_start ../backend/wayland/backend.c:55
#7 0x7f55cc1bad55 in wlr_backend_start ../backend/backend.c:35
#8 0x7f55cc1d67a0 in multi_backend_start ../backend/multi/backend.c:24
#9 0x7f55cc1bad55 in wlr_backend_start ../backend/backend.c:35
#10 0x40ba8a in server_run ../sway/server.c:136
#11 0x40ab2f in main ../sway/main.c:432
#12 0x7f55cb97318a in __libc_start_main ../csu/libc-start.c:308
Emitting the close event needs to happen before
container_output_destroy, because container_output_destroy sets the
sway_output to NULL and sway_output is used in IPC.
We were freeing the sway_output immediately upon disconnect which left
a dangling pointer in the output's container. It then tried to use the
pointer when the container is freed.
We don't need to store the sway_output in an output's container which is
destroying, so the fix is to set the pointer to NULL and remove the use
in container_free.
Also added an arrange when the output is disconnected for good measure.
Prompts e.g. authentication request from firefox-wayland ought to be
floating.
This is a bit coarse but just fixed size is not enough, here is what
firefox does:
[1285461.363] -> xdg_wm_base@18.get_xdg_surface(new id xdg_surface@68, wl_surface@71)
[1285461.508] -> xdg_surface@68.get_toplevel(new id xdg_toplevel@67)
[1285461.571] -> xdg_toplevel@67.set_parent(xdg_toplevel@37)
[1285461.630] -> xdg_toplevel@67.set_title("Authentication Required")
[1285461.736] -> xdg_toplevel@67.set_app_id("firefox")
...
[1285476.549] xdg_toplevel@67.configure(0, 0, array)
...
[1285502.080] -> xdg_toplevel@67.set_min_size(299, 187)
[1285502.140] -> xdg_toplevel@67.set_max_size(1920, 32767)
This can also be observed with e.g. the open window of gedit
(gedit->open->other documents)
It happened when a view is a grandchild or deeper of the workspace, is
fullscreen, and unmaps. The workspace would not be included in the
transaction and its pointer to the fullscreen view was left dangling.
container_destroy was calling container_reap_empty, which calls
container_destroy and so on. Eventually the original container_destroy
would return a NULL pointer to the caller which caused a crash.
This also fixes an arrange on the wrong container when moving views in
and out of stacks.
if src is NULL due to a previous error we cannot use it in the command
result string.
Moreover if `src` points to `p.we_wordv[0]` we cannot use it after
`wordfree(&p)` in the command result string.
Bonus feature: If there was an error accessing the file, the string
rapresentation of the error is now included in the command result
string.
Some operations during backend creation (e.g. becoming DRM master)
require CAP_SYS_ADMIN privileges. At this point, sway has dropped them
already, though. This patch splits the privileged part of server_init
into its own function and calls it before dropping its privileges.
This fixes the bug with minimal security implications.
* Ensure that modifier keys are identified even when the next key does
not produce a keysym. This requires that modifier change tracking
be done for each sway_shortcut_state.
* Permit regular and --release shortcuts on the same key combination.
Distinct bindings are identified for press and release cases; note
that the release binding needs to be identified for both key press
and key release events.
* Maintain ascending sort order for the shortcut state list, and keep
track of the number of pressed key ids, for simpler (and hence
faster) searching of the list of key bindings.
* Move binding duplicate detection into get_active_binding to avoid
duplicating error messages.
Sort the list comprising the set of keys for the binding in ascending
order. (Keyboard shortcuts depend only on the set of simultaneously
pressed keys, not their order, so this change should have no external
effect.) This simplifies comparisons between bindings.
* The arrange_foo functions are now replaced with arrange_and_commit, or
with manually created transactions and arrange_windows x2.
* The arrange functions are now only called from the highest level
functions rather than from both high level and low level functions.
* Due to the previous point, view_set_fullscreen_raw and
view_set_fullscreen are both merged into one function again.
* Floating and fullscreen are now working with transactions.
