- When policies are allocated, the ipc target path goes
through symlink resolution. The result is used as
the canonical for matching pids to policies at runtime.
In particular, this matches up with the target of
the `/proc/<pid>/exe`.
- There's a possible race condition if this isn't done
correctly, read below.
Originally, validate_ipc_target() always tried to resolve
its argument for symlinks, and returned a parogram target string
if it validates. This created a possible race condition with
security implications. The problem is that get_feature_policy()
first independently resolved the policy target in order to check
whether a policy already exists. If it didn't find any, it called
alloc_feature_policy() which called validate_ipc_target() which
resolved the policy target again. In the time between the two
checks, the symlink could be altered, and a lucky attacker could
fool the program into thinking that a policy doesn't exist
for a target, and then switch the symlink to point at another file.
At the very least this could allow him to create two policies
for the same program target, and possibly to bypass security
by associating the permissions for one target with another,
or force default permissions to apply to a target for which
a more specific rule has been configured. So we don't that.
Instead, the policy target is resolved once and that result is
used for the rest of the lookup/creation process.
In i3 every mark is unique and one mark cannot be used in more than one
window, sway behavior has been amended to match this.
`swaymsg -t get_marks` will now return an array of all marks used in sway.
See #98
- Moved ``<sys/capability.h>`` include inside `__linux__` guard,
because all uses are similarly guarded.
- <sys/capability.h> is part of an optional devel package, at least
in fedora. CMake now explicitly checks that libcap devel files
are available.
- Added libcap to the list of install packages in .travis.yml, to
make the dependency explicit. travis-ci installs the package by
default, which is why this hasn't surfaced previously.
This commit changes how commands decide what container to act on.
Commands get the current container though `current_container`, a global
defined in sway/commands.c. If a criteria is given before a command,
then the following command will be run once for every container the
criteria matches with a reference to the matching container in
'current_container'. Commands should use this instead of
`get_focused_container()` from now on.
This commit also fixes a few (minor) mistakes made in implementing marks
such as non-escaped arrows in sway(5) and calling the "mark" command
"floating" by accident. It also cleans up `criteria.c` in a few places.
This commit adds three commands to sway: `show_marks`, `mark` and
`unmark`. Marks are displayed right-aligned in the window border as i3
does. Marks may be found using criteria.
Fixes#1007
Fixes#1120
When the parent of a view is C_WORKSPACE and the movement direction
is either MOVE_PREV or MOVE_NEXT, the code would attempt to move the
views to the next output, but swayc_adjacent_output can't accept
non-directional movement commands and causes undefined behaviour and
a segfault.
If the code is simply skipped, we end up in an infinite loop.
Instead, we can allow containers whose parent is a C_WORKSPACE take the
path that handles MOVE_PREV and MOVE_NEXT, which behaves as you would
expect.
I'm not certain that this fix is entirely correct as the desired behaviour
of move_container is not very well defined, but it seems to work.
Hardcoding it to L_HORIZ does not make sense to me,
as you get the unexpected behaviour that windows will be
arranged horizontally until you switch the layout.
As best I can tell this todo was intended to add workspace movement to
the given output with the `workspace <ws> output <op>` command, but i3
does not behave this way.
This commit allows unquoted spaces in worspace names in order to keep
compatability with i3. The names _must not_ contain the string "output"
which is documented in 'sway.5' because how sway detects the `move
<workspace> output <output>` command. Also I documented that "number"
may be used before the worspace name without affecting how the name is
evaluated.
This commit lets the 'move' command apply to floating containers as well
as tiled ones. The command may be appended with a number of pixels and
then optionally the string `px` (like '10 px') in order to move the
container more or fewer than the standard ten pixels.
Previous implementation would not preserve dimension of groups
along the major axis. This should avoid weird behavior when
using container motion commands.
This fixes issue #733. Now if the user focuses output right but is at
the rightmost monitor, the focus will wrap the the leftmost monitor.
This commit adds a new function, swayc_opposite_output, which selects
the opposite output given a position and a direction. Now, when calling
output_by_name, we first check if there is an adjacent output to switch
to. If that fails, we call swayc_opposite_output to handle wrapping.
- "layout auto_left|auto_xxx" are now "layout auto xxx"
- "layout incmaster <n>" is now "layout auto master [set|inc] <n>"
- "layout incncol <n>" is now "layout auto ncol [set|inc] <n>"
- prior to this modification, the requested pixels were added/removed
to both edges of the modified container. To preserve sizes,
only half the pixels should be added/removed to each edge.
- added L_AUTO_FIRST/LAST instead of using explicit layouts.
- when switching between auto layout that don't share the same major axis, invert the
width/height of their child views to preserve their relative proportions.
Don't switch the internal tracking of focus to the swaylock surface,
to allow for switching back to the previously active window (or the
currently active window, if some new process changed).
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
When destroying lock surfaces, we really should only unlock a
desktop_shell if the set of lock surfaces has dropped to zero (since
callers need to do a set_lock_surface for every output).
Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
Exherbo installs architecture dependent data in a different place than architecture
independent data. More concretely: binaries go in /usr/$chost/{bin,lib},
data goes in /usr/share and configs in /etc, /etc is already configurable
through CMAKE_INSTALL_FULL_SYSCONFDIR but the datadir was not. This
patch fixes it so that things can be pushed in the right places.
- Make sure CMake always finds absolute paths for Cairo, Pango and GdkPixbuf
- Add forgotten json-c include path to swaymsg/CMakeLists.txt
- Disable -Werror because of assert warnings
- Add correct /proc/pid/file path for FreeBSD
- Use libepoll-shim on FreeBSD
- Only use Linux capabilities on, well, Linux
For workspace containers, swayc_change_layout also changes ->layout alongside
->workspace_layout when it's a sensible thing to do. There is an additional test
for 'layout toggle' command which ensures that containers will be tiled
horizontally after toggling from tabbed or stacked.
If workspace layout is set to tabbed or stacked, its C_VIEW children
should get wrapped in a container. Alongside that, move_container was
modified to retain previous functionality.
LD_PRELOAD enables keyloggers to easily be made. This solution isn't
perfect - really a secure system wouldn't have LD_PRELOAD at all. It was
a stupid idea in the first place.
