When the theme variables were moved to the backend instead of `rc.lua`,
some magic was added to disable them if the user set the border. However,
some undocumented `awful.placement` code also set them and turned off
the theme variables. So it worked *once* then stopped working.
Previously, it only added 1 of the 2 sides of the relevant margins
to the struct size. For example, if the position was "top", then
only the top margin was added, not the bottom one.
Without this change, if a composed placement was used, it would leak
the "override_geometry" in the args, which would be kept cached during
the next composed node. If that node had no override of its own, it
would use the previous one by accident.
In practice, it means it was impossible to resize a `wibar` manually.
The `wibar` would always restore itself to the height (or width, for
vertical ones) it previously had.
The function returns the geometry of the parent object. As it was used,
it was passed the parent object directly. The means the parent geometry
of the parent object was used instead of the geometry of the parent
object. This worked "fine" as long as the mouse was in the same screen,
but it was just hiding the bug.
The default `rc.lua` was using the focus/unfocus signals to set
the border color along with `awful.rules`. This logic block was
no longer aligned with the rest of `rc.lua` since it was
the only place where `beautiful` variables where only used by
`rc.lua`.
On top of this, the new request handler also has extra contexts
for the urgent and floating/maximixed use cases. So it can be used
by themes to implement much smarter borders than just focus based
ones. They were previously limited by the fact most of the
(un-monkey-patchable) logic was in `rc.lua`.
Note that this commit also shuffle the awful.rules order between
the titlebar and the border and changes the tests accordignly.
After some consideration, I came to the conclusion the previous
behavior was bogus and the fact that the placement tests required
to know about the titlebar height is simply a proof of that. The
change was required in this commit because since the border is no
longer in the default rules, a new buggy edge case surfaced.
`awful.client` was only used once and the function exists as a
screen method. This will help untangle the dependencies a bit...
To tangle them even more after that...
The awful.placement.no_overlap function always looked at the currently
visible clients when placing a new client. This produced a confusing
result when using awful.rules or the sn_rules argument of awful.spawn to
place the client on an unselected tag (the client was placed as if it
would be placed on a currently selected tag; if multiple clients were
placed on the same unselected tag, in many cases they were placed at the
same position, overlapping each other).
Make awful.placement.no_overlap check tags of the placed client and
handle the case of placement on an unselected tag in a more useful way:
- If the client is sticky or at least one of the client tags is
selected, keep the previous behavior: avoid overlap with all other
floating clients which are currently visible, and use the currently
active layout to determine the floating status.
An explicit check based on `c:tags()` is made instead of using
`c:isvisible()`, so that the previous behavior is kept even if the
client is hidden or minimized for some reason.
- If all client tags are unselected, avoid overlap with all other
floating clients which either are sticky or share at least one tag
with the placed client, and use the layout of the first tag of the
placed client to determine the floating status.
Signed-off-by: Sergey Vlasov <sigprof@gmail.com>
This way their name doesn't get mangle by the broken magic. It will also
eventually allow to `error()` in the template when the implicit
`@function` is used.
This commit also fixes a large number of issues found while
proof-reading everything.
Given noone understand this code, this will prevent some semi likely
regressions from going unnoticed. The main risk is the shims not
producing the exact same results as the real implementation and
cause different code paths to be taken.
As of this commit, both the "real" and "shim" implementation were given
the same set of tests with print() at every step of next_to. The
resulting log was then checksummed to ensure both are identical.
The awful.placement.no_offscreen function did not work properly when
composed with other placement functions; in particular, the default
configuration (awful.placement.no_overlap+awful.placement.no_offscreen)
was broken. The compose function sets args.pretend=true and puts the
result of the previous placement function into args.override_geometry
before calling the next placement function, but no_offscreen did not use
args.override_geometry, therefore the result of the previous placement
function was discarded.
All other placement functions use `geometry_common(c, args)` to get the
current client geometry; `area_common(c)` should be used only when
getting geometry of other clients.
This change also fixes the problem with margin handling (adding margins
should not affect the window size, only the window position should
change); the test output which was adjusted in commit 0275d3537d
is adjusted again to account for this change.
Signed-off-by: Sergey Vlasov <sigprof@gmail.com>
The awful.placement.no_overlap function was adding the window border
width to the client width and height (this is performed in
area_common(), which is called by geometry_common()), but did not
reverse this operation by calling remove_border() before returning the
final geometry; because of this, using no_overlap resulted in increasing
the window width and height by 2*border_width.
The bug was probably introduced in commit ebcc19844e (before
that commit no_overlap changed the window position directly instead of
relying on the new placement infrastructure), but was not noticed
because of other problems (e.g., in the default configuration the result
of no_overlap was overridden by the buggy no_offscreen).
Signed-off-by: Sergey Vlasov <sigprof@gmail.com>
Once upon a time, 4b9584fdb1 already fixed this problem: We have to
set the border width to zero before applying the new geometry to the
client, because changing the border width makes the client move
according to its gravity.
Then came e54387904b and made this code use awful.placement instead
of just fullscreening the client itself (without explaining why in the
commit message!). After this commit, the border width was just ignored
and left as-is. This was then fixed in 0bf8bb6a64 (no idea which
callback the commit message refers to, the old code was basically just
c.border_width=0, c:geometry(screen_geo)). However, now the border width
was again changed after the geometry and the bug that was fixed by
4b9584fdb1 was back.
This commit fixes this regression again by making sure that the border
width is set to zero before the geometry is set. This becomes slightly
more complicated, because now it is also awful.placement's job to
restore the old border width.
This is why this commit adds a new option to awful.placement so that it
sets the border width to zero after creating its memento of the old
border width.
Fixes: https://github.com/awesomeWM/awesome/issues/1607
Signed-off-by: Uli Schlachter <psychon@znc.in>
* Move table functions out of awful.util into new gears.table
* travis: Use v9999 prefix for full requests
Make sure no newly deprecated functions are used
* Move all `awful.util.table.*` calls to `gears.table.*` calls
Move table test functions from awful/util_spec to new gears/table_spec
Change awful.util.subsets call to gears.math.subsets in awful/key.lua
It does not provide much value. The version number is already known to
ldoc globally in the "description" variable.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This commit add the last placement function imported from the
Radical module.
It allows to place a wibox/client next to another object. It tries
to find the best fit. It also support wibox widgets.
This is intended for tooltips and menus, but can also be used in
`awful.rules` to place the new client as close as possible to the
focused one without overlap.
The requirement to call add_signal() was added to catch typos. However, this
requirement became increasingly annoying with property::<name> signals and e.g.
gears.object allowing arbitrary properties to be changed.
All of this ended up in a single commit because tests/examples fails if I first
let add_signal() emit a deprecation warning.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Emmanuel Lepage Vallee
sarubo
Elyes HAOUAS
James Reed
actionless
Sergey Vlasov
Anatolii Aniskovych
Uli Schlachter
Kevin Zander
Daniel Hahler
cmertz
Uli Schlachter