When unmanaging a window we need to reparent it back to the root
window. With this patch we do so at the current position of the
container such that we don't throw away the geometry position.
This fixes cases where clients withdraw a window and reparent it
later on, expecting to keep the same geometry (in particular dock
clients) but then end up on a wrong output.
fixes#3819
I had a dilemma about the behaviour here:
1. Prohibit focus leaving the workspace in any case unless if
get_tree_next's initial argument is a workspace. This is what this
commit does (also i3-cycle).
2. Leave the workspace if no warp is possible (eg workspace with single
container or `focus right` with `V[a b c*]`).
Fixes#2180
- Makes `tree_next` not recursive.
- Adds `focus next|prev [sibling]` command. See (1.) and (2.) in
https://github.com/i3/i3/issues/2587#issuecomment-378505551 (Issue also
requests move command, not implemented here).
- Directional focus command now supports command criteria.
Wrapping is not implemented inside a floating container. This was also
true before the refactor so I am not changing it here.
This commit should fix "move con to parent" trick (see below) in the
case when con->parent->parent is a workspace.
The trick:
mark _a, focus parent, focus parent, mark _b,
[con_mark=_a] move window to mark _b, [con_mark=_a] focus
The trick got broken in commit 626af81232
in order to fix an i3 crash (#2003). Reverting said commit fixes the
trick. The crash is caused by the fact that empty workspace isn't
considered a split (checked in src/con.c:1324), so the moved window ends
up as a sibling of the target workspace, not as its child.
Move drag_pointer() and related definitions from floating.c to new file
drag_pointer.c since it's applicable not only to floating windows but
also to resizing of tiled windows.
Some apps including XTerm start with a WM_CLIENT_LEADER property
containing their own window ID. Before this commit, i3 tried to center
such windows onto itself and did it wrong since
`leader->rect == {0,0,0,0}` at this moment.
The first affected commit is 128122e766,
however, before it such windows already was misplaced, but got sanitized
afterward [1].
[1]: 8a3ef3a81b/src/floating.c (L329-L335)Fixes#3606
Fixes#3785 -- the issue where the Travis build failed because of gcc's
-Werror=address-of-packed-member.
Adds an equality function to avoid relying on memcmp().
This fixes the case where moving a container to a scratchpad hidden
container via a mark would cause the container to be tiling on the
__i3_scratch workspace. This still moves the container to the
__i3_scratch workspace, but properly adds it to the scratchpad so that
it becomes usable instead of requiring criteria to regain access to.
This is achieved by retaining the IPC connection which is sending the restart
command across the restart.
This is the cleaner fix for https://github.com/i3/go-i3/issues/3fixes#3565
The testcase is changed because it was actually incorrect. Easy to
verify because:
> _NET_CURRENT_DESKTOP
> …
> The index of the current desktop. This is always an integer between 0
> and _NET_NUMBER_OF_DESKTOPS - 1.
Fixes#3696.
Also updates the viewports.
Finally, fixes an issue with _NET_CURRENT_DESKTOP not being updated
after a workspace rename. Example:
- workspaces 1, 2, 3
- rename workspace 1 to 5
- All workspaces changed their index but _NET_CURRENT_DESKTOP was not
updated
- manage_window: log the window in the start of the function so that the
reader knows what the rest of the messages refer to, even if the
function exits prematurely.
- con_is_floating: Message is spammy
- Having both parse_configuration and parse_file is excessive now
- We detect if we are parsing only by checking if conn is NULL, not with
use_nagbar
- font.pattern needs to be set to NULL because it is freed in
free_font()
Fixes#3660
Closes#3635.
Probably the bug can still happen when a tree_close_internal happens
inside a workspace_show but modifying the code to avoid them seems to
not be worth it.
- The result from con_get_output was always not NULL because
con_get_output asserts so
- get_output_by_name should always be able to get an output from the
corresponding container
- workspace_move_to_output doesn't return bool anymore since it can't
fail
This is a regression from bce088679.
An other way to fix this would be to concatenate strings inside the
strtok loop when an output starts with a double quote but I'd rather
let the parser do the word splitting.
Fixes#3646
This fixes a crash which I could not reproduce in a testcase with reasonable effort, but the user reported the fix works.
Compare with src/x.c:946.
fixes#3554fixes#3645
Fixes#3595
Like the issue mentions:
> instead of the newly created workspace (not referenced by variable
> here) the `"init"` event is fired with the current workspace (`ws`).
Plus, there was another issue where duplicate workspace init events
where being sent because of workspace_get().
304-ipc-workspace-init.t: Subtest "move workspace to output" fails with
current next.
Fixes#3631
No event was being sent here:
2d6e09a66a/src/randr.c (L487)
533-randr15.t: I confirmed that SKIP still works if the xrandr command
fails.
Added test fails with current next.
- __lsan_do_leak_check() will terminate the process, so move it to the
end of the function.
- ev_loop_destroy() must be called after ipc_shutdown() because the
latter calls ev_ functions.
Fixes#3599
I was able to reproduce #3579 in Linux by running:
`sudo sysctl net.core.wmem_default=10000`
If a subscription message was too big to be sent at once, it was
possible to break a client by sending a reply to an other message sent
by the client. Eg:
- Write 8192 out of 11612 bytes of a workspace event.
- Blockingly write the reply to a workspace change message.
- Write the rest 3420 bytes of the workspace event.
This commit fixes this by utilizing the ipc queue for all types of
writes.
ipc_receive_message can only be called from a callback started in
ipc_new_client. This callback uses the same file descriptor with the
client also created in ipc_new_client. When the client is deleted, the
read callback is now also stopped. Thus, we can assume that whenever
ipc_receive_message is called, the corresponding client should still
exist.
- ipc_client now contains pointers to both write and read watchers. When
freed, a client will stop both of them.
- IPC_HANDLERs now work with ipc_clients instead of fds.
Fixes#3579.
errx() already appends \n internally. "\n" in the error message will
result in a blank line after the message. die() is just a wrapper around
errx() so it receives the same treatment.
After 204eefc. Alternative fix:
diff --git a/src/floating.c b/src/floating.c
index f5c61782..6dd79668 100644
--- a/src/floating.c
+++ b/src/floating.c
@@ -954,7 +954,7 @@ bool floating_reposition(Con *con, Rect newrect) {
con->scratchpad_state = SCRATCHPAD_CHANGED;
/* Workspace change will already result in a tree_render. */
- if (!reassigned) {
+ if (!reassigned && workspace_is_visible(con_get_workspace(con))) {
render_con(con);
x_push_node(con);
}
but I don't think that the extra complexity is worth it.
Change in handlers.c because of d2d6d6e0 where the bug also appears.
Fixes#3567
on_remove_child calls tree_close_internal which calls tree_render and
the tree is in an invalid state if con->parent still points to the old
parent.
Fixes#3556
Using the wrong X11 connection breaks the libev event handling model:
xcb_flush() must be called immediately before handing control to libev.
Before this fix:
1. xcb_prepare_cb would read and flush conn
2. restore_xcb_prepare_cb would read and flush restore_conn,
BUT also inadvertantly call xcb_flush(conn), resulting in new
events being filled into the XCB event queue
3. libev waits for new events
4. after 1 minute, libev times out and the events are processed
Diagnosed using strace on testcases/complete-run.pl.
related to commit 0d8b6714e3
related to #3510
Basic idea: if the window has a shape, set the parent container shape as
the union of the window shape and the shape of the frame borders.
Co-authored-by: Uli Schlachter <psychon@znc.in>
- ICCCM says: > If a base size is not provided, the minimum size is to
be used in its place and vice versa.
i3 didn't obey the "vice versa" part. Min size and base size are both
saved without replacements in window_update_normal_hints,
floating_check_size makes the needed replacements if either was not
provided.
- Aspect ratio is now saved correctly in manage_window because
window_update_normal_hints is called.
- i3 didn't save the aspect ratio if the window conformed the given
aspect ratio range when handle_normal_hints was called. If the window
was resized to a size outside of the given bounds, i3 didn't correct it.
- Aspect ratio now affects only tiling windows, like the rest of the
normal size hints
- The aspect ratio calculation is now done without a loop
A real life example of how these changes affect the workflow:
An mpv window, when playing a video, sets its min == max aspect ratio
during mapping. i3 ignored these hints. When resized, the window's
aspect ratio was not preserved. With this commit, resizing floating mpv
windows will always preserve the aspect ratio.
Previously, it first calculated one of the containers' next percentage, and then subtracted the previous percentage to find the actual change.
Now it directly calculates the change, and subtracts and adds the change to the two affected containers.
Added util function con_rect_size_in_orientation.
Removed px_resize_to_percent; inlined, using con_rect_size_in_orientation.
Also, prematurely return when pixel diff is 0, as no action is necessary.
This is related to [this issue on i3-gaps](https://github.com/Airblader/i3/issues/247).
Michael Stapelberg